# AI for science

AI for science is research in artificial intelligence, including scientific machine learning, to accelerate scientific discovery.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## DeepQuest: From Knowledge Gaps to Scientific Discovery in Alzheimer's Disease

DevFeed: [DeepQuest: From Knowledge Gaps to Scientific Discovery in Alzheimer's Disease](<https://devfeed.tech/articles/deepquest-from-knowledge-gaps-to-scientific-discovery-in-alzheimer-s-disease-145661.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26343>)

Author: Chunbo Jiang, Yifei Chen, Beining Wu, Fuyou Mao, Weihong Chen, Enyu Bao, Mingxuan Liu, Feiwei Qin, Qiyuan Tian

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Data Science](<https://devfeed.tech/topics/data-science.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [audit](<https://devfeed.tech/tags/audit.md>), [automated](<https://devfeed.tech/tags/automated.md>), [code](<https://devfeed.tech/tags/code.md>), [comparisons](<https://devfeed.tech/tags/comparisons.md>), [contribution](<https://devfeed.tech/tags/contribution.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [data](<https://devfeed.tech/tags/data.md>), [dimension](<https://devfeed.tech/tags/dimension.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [disease](<https://devfeed.tech/tags/disease.md>), [domain](<https://devfeed.tech/tags/domain.md>), [experimental](<https://devfeed.tech/tags/experimental.md>), [experimentation](<https://devfeed.tech/tags/experimentation.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gap](<https://devfeed.tech/tags/gap.md>), [general](<https://devfeed.tech/tags/general.md>), [independent](<https://devfeed.tech/tags/independent.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>)

### AI overview

DeepQuest turns unresolved questions from Alzheimer's disease reviews into testable hypotheses, screens them for data feasibility, and prioritizes them with groupwise Elo ranking. Its iterative experiments are audited separately for reliability and contribution relative to two frozen literature corpora. On OASIS-3, it generated 42 feasible hypotheses and beat the strongest baselines in 14 of 15 evaluator-dimension comparisons across three language-model reviewers. The framework's audit-assessed scientific finding rate was 95.24%, exceeding general-purpose executors and domain-specific research agents and remaining stable across literature corpora and evaluator model families. Reported findings address neuropsychiatric symptoms and dementia risk, longitudinal amyloid spatial stability, and neocortical tau-associated clinical impairment; key associations were supported in independent ADNI and AIBL cohorts.

### Source excerpt

arXiv:2611.26343v1 Announce Type: new Abstract: Automating experiments does not necessarily advance scientific understanding. Research agents must identify meaningful knowledge gaps, determine what available data can resolve, and assess whether their findings advance existing knowledge. We introduce DeepQuest, a knowledge-gap-driven AI scientist framework for Alzheimer's disease research. DeepQuest transforms unresolved questions from domain reviews into traceable, testable hypotheses and prioritizes them through data feasibility screening and groupwise Elo ranking. It couples iterative experimentation with two-level auditing that separates experimental reliability from knowledge contribution relative to two frozen literature corpora. On OASIS-3, DeepQuest produces 42 feasible hypotheses and outperforms the strongest baselines in 14 of 15 evaluator-dimension comparisons across three language-model reviewers. The full framework achieves an audit-assessed scientific finding rate of 95.24%, outperforming both general-purpose executors and domain-specific research agents. This scientific finding rate remains stable across literature corpora and evaluator model families. Representative findings concern neuropsychiatric symptoms and dementia risk, longitudinal amyloid spatial stability, and neocortical tau-associated clinical impairment, with key associations supported in independent ADNI and AIBL cohorts. These results show how aligning knowledge gaps, data feasibility, and evidence auditing can turn automated research into traceable, evidence-supported scientific discovery. Our source code is available at https://anonymous.4open.science/r/DeepQuest-3066/

## Why Do LLMs Help with Scientific Misconduct Requests and What to Do About It?: Increasing the Trustworthiness of AI for Science

DevFeed: [Why Do LLMs Help with Scientific Misconduct Requests and What to Do About It?: Increasing the Trustworthiness of AI for Science](<https://devfeed.tech/articles/why-do-llms-help-with-scientific-misconduct-requests-and-what-to-do-about-it-increasing-the-trustworthiness-of-ai-for-science-145523.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25492>)

Author: Almene De Meran Meguimtsop, Daniel E. Acuna

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [LLM security](<https://devfeed.tech/topics/llm-security.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [llms](<https://devfeed.tech/tags/llms.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [prompts](<https://devfeed.tech/tags/prompts.md>), [research-integrity](<https://devfeed.tech/tags/research-integrity.md>), [safety](<https://devfeed.tech/tags/safety.md>), [science](<https://devfeed.tech/tags/science.md>)

### AI overview

Across eight LLMs tested on 1,350 SciIntBench prompts, models often recognized scientific misconduct requests but rarely refused them: they identified misconduct in 91% of cases when asked directly, yet refused in 34%. Request framing was associated with refusal; changing instructions into questions about permissibility raised refusal from 52% to 88%. Two tested system-prompt changes--asking models to assess a request before answering and reminding them of responsible conduct of research principles--raised refusal on adversarial requests from 58% to 87% and 86%, respectively. The abstract reports these as study findings, not evidence that the changes prevent misconduct in practice.

### Source excerpt

arXiv:2611.25492v1 Announce Type: new Abstract: Large language models (LLMs) support scientific research, but some of that assistance can also undermine research integrity. Existing safety evaluations in science often measure whether models refuse misconduct requests, but rarely explain whether they even recognize the problem or why they comply. We investigate this gap across 8 LLMs using SciIntBench, an existing benchmark of 1,350 prompts about scientific research integrity. We examine the reasons behind compliance in four steps. First, when asked directly whether fulfilling a request would involve misconduct, models overwhelmingly identify such requests, even in the Most Covert cases (91%), while they refuse to carry out the request in only 34% of cases. Second, we examine which request characteristics are associated with refusal using features that capture how the request is framed, how much the assistant has been told about what it is helping to hide, and how the goal and the user's stake are stated, among others. Third, we estimate that the most significant effect was to rephrase requests as questions about permissibility rather than instructions, which increased refusal from 52% to 88%. Finally, we test two system-prompt fixes: asking models to assess the request before responding and reminding them of responsible conduct of research principles. These changes increase refusal rates on adversarial requests from 58% to 87% and 86%, respectively. Our findings show that models recognize wrongful requests but still help users violate integrity. Understanding why these failures occur and testing ways to reduce them offer promising paths toward more trustworthy AI for science.

## MatVerse: A Foundation Model for Materials Science

DevFeed: [MatVerse: A Foundation Model for Materials Science](<https://devfeed.tech/articles/matverse-a-foundation-model-for-materials-science-145560.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25651>)

Author: Xuanyu Zhu, Qian Tan, Bo Chen, Jiaqing Xie, Yang Han, Mei Wu, Wanhao Liu, Xuanze Lin, Yuchen Zhou, Shijie Wang, Biqing Qi, Shixiang Tang, Wenwei Zhang, Tianfan Fu, Kai Yu, Lu Chen, Yuqiang Li, Xin Che

Published: 2026-11-20T05:00:00Z

Content type: news

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Materials science and engineering](<https://devfeed.tech/topics/materials-science-and-engineering.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [post-training](<https://devfeed.tech/topics/post-training.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-assisted](<https://devfeed.tech/tags/ai-assisted.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [cond-mat-mtrl-sci](<https://devfeed.tech/tags/cond-mat-mtrl-sci.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [foundation-model](<https://devfeed.tech/tags/foundation-model.md>), [inference-efficiency](<https://devfeed.tech/tags/inference-efficiency.md>), [materials](<https://devfeed.tech/tags/materials.md>), [materials-science](<https://devfeed.tech/tags/materials-science.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [post-training](<https://devfeed.tech/tags/post-training.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [vision](<https://devfeed.tech/tags/vision.md>)

### AI overview

MatVerse is a materials-science foundation model built on Intern-S2-Preview and specialized through large-scale domain post-training. Its multimodal corpus covers materials knowledge, scientific reasoning, quantitative problems, structures, and scientific interpretation; data preparation combines teacher-guided reasoning distillation, synthesized problems, consistency filtering, and retrieval-based multimodal construction. The report says MatVerse improves over the base model on OmniMatBench, MatSciBench, and MaCBench, and improves inference efficiency on calculation-intensive problems. The excerpt gives no numerical results or detailed benchmark comparisons.

### Source excerpt

arXiv:2611.25651v1 Announce Type: new Abstract: Materials science involves highly heterogeneous scientific information, including textual knowledge, molecular and crystal structures, experimental characterization data, microscopy images, and structure-property relationships. Developing a general-purpose foundation model capable of understanding and reasoning over these diverse materials representations remains challenging. In this technical report, we introduce MatVerse, a materials foundation model built upon Intern-S2-Preview and further specialized through large-scale materials-domain post-training. We construct a large-scale multimodal training corpus covering materials knowledge, scientific reasoning, quantitative problem solving, structural understanding, and multimodal scientific interpretation. To improve domain adaptation, we develop a unified data construction pipeline combining teacher-guided reasoning distillation, scientific problem synthesis, consistency-based filtering, and retrieval-based multimodal data construction. We evaluate MatVerse on three representative materials-science benchmarks: our human-calibrated OmniMatBench, MatSciBench, and MaCBench. Results demonstrate consistent improvements after materials-specific post-training over the base model on all three benchmarks, alongside improved inference efficiency on calculation-intensive problems. These results show that targeted domain-specific post-training can effectively enhance scientific reasoning, materials knowledge understanding, and multimodal scientific capabilities of foundation models. MatVerse provides a unified language-and-vision interface for materials science and serves as a foundation for future AI-assisted materials research workflows.

## Steering Automated Scientific Discovery via Audited Epistemic Exploration

DevFeed: [Steering Automated Scientific Discovery via Audited Epistemic Exploration](<https://devfeed.tech/articles/steering-automated-scientific-discovery-via-audited-epistemic-exploration-145483.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25214>)

Author: Fangxu Yu, Tao Feng, Xuying Ning, Haolin Yang, Jiaxuan You, Tianyi Zhou

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [automated](<https://devfeed.tech/tags/automated.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [exploration](<https://devfeed.tech/tags/exploration.md>), [framework](<https://devfeed.tech/tags/framework.md>), [hypothesis](<https://devfeed.tech/tags/hypothesis.md>), [investigation](<https://devfeed.tech/tags/investigation.md>), [llm](<https://devfeed.tech/tags/llm.md>), [reductions](<https://devfeed.tech/tags/reductions.md>), [research](<https://devfeed.tech/tags/research.md>), [results](<https://devfeed.tech/tags/results.md>), [revisions](<https://devfeed.tech/tags/revisions.md>)

### AI overview

EpiCompass is a plug-and-play scaffold for scientific research agents that tracks hypotheses, evidence, beliefs, and revisions in an epistemic tree. It reassesses explored hypotheses against accumulated evidence, uses unresolved conflicts and unexplored directions to generate candidates, and selects them for anticipated epistemic gain--either refining existing hypotheses or expanding the hypothesis space. It also checks final conclusions against the evidence. Across three discovery benchmarks and three LLM backbones, the abstract reports consistent gains over baselines and reductions in unsupported claims and failures to revise beliefs.

### Source excerpt

arXiv:2611.25214v1 Announce Type: new Abstract: Scientific discovery requires revisiting the hypotheses in light of new evidence and identifying promising directions beyond those already explored. However, research agents often make claims that are not supported by their observations, explore a narrow hypothesis space, and retain beliefs despite recognizing conflicting evidence. These failures motivate the search for an explicit epistemic state that captures both beliefs about existing hypotheses and directions for further exploration. To this end, we propose Epistemic Compass (EpiCompass), a plug-and-play scaffold that maintains this state through an epistemic tree linking hypotheses, evidence, beliefs, and revisions. As an investigation progresses, EpiCompass reassesses the hypotheses explored against accumulated evidence and uses unresolved conflicts and unexplored directions to guide hypothesis generation. It selects candidates by anticipated epistemic gain, accounting for both refinement of existing hypotheses and expansion of the hypothesis space, and verifies the final conclusion against the accumulated evidence. Experimental results across three scientific discovery benchmarks and three LLM backbones demonstrate that EpiCompass consistently outperforms baselines, with further analyses showing reductions in errors of unsupported claims and failures to revise beliefs.

## Medical AI startup Boltz raises $110M to speed up small molecule research

DevFeed: [Medical AI startup Boltz raises $110M to speed up small molecule research](<https://devfeed.tech/articles/medical-ai-startup-boltz-raises-110m-to-speed-up-small-molecule-research-144377.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/19/medical-ai-startup-boltz-raises-110m-to-speed-up-small-molecule-research/>)

Author: Maria Deutscher

Published: 2026-11-19T21:17:08Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-startup](<https://devfeed.tech/tags/ai-startup.md>), [boltz](<https://devfeed.tech/tags/boltz.md>), [boltzmol-1](<https://devfeed.tech/tags/boltzmol-1.md>), [boltzprot-1](<https://devfeed.tech/tags/boltzprot-1.md>), [funding-round](<https://devfeed.tech/tags/funding-round.md>), [investment](<https://devfeed.tech/tags/investment.md>), [medical](<https://devfeed.tech/tags/medical.md>), [medical-ai](<https://devfeed.tech/tags/medical-ai.md>), [news](<https://devfeed.tech/tags/news.md>), [protein](<https://devfeed.tech/tags/protein.md>), [raises](<https://devfeed.tech/tags/raises.md>), [research](<https://devfeed.tech/tags/research.md>), [small-molecule](<https://devfeed.tech/tags/small-molecule.md>), [speed](<https://devfeed.tech/tags/speed.md>), [startup](<https://devfeed.tech/tags/startup.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [top-story-6](<https://devfeed.tech/tags/top-story-6.md>)

### AI overview

Boltz raised $110 million in a Series A led by Bessemer to expand staff and computing capacity. Its BoltzMol-1 model narrows small-molecule screening from hundreds of thousands of candidates to a few dozen promising ones; Boltz says it found small-molecule hits for six of 10 challenging targets. These hits are research clues, not drug discoveries. Researchers can constrain searches to compounds they can access or broaden them when procurement is not a barrier. BoltzProt-1 generates protein designs from hit data and evaluates their potential with more direct metrics; the company reports higher success than several competing models in internal tests. Boltz sells both models through a token-billed API, serves six of the 20 largest pharmaceutical companies, and says customer token use has grown 50-fold since the models launched in June. The article notes competition from biology-focused models at Anthropic and OpenAI and says Boltz is training a new generation of algorithms.

### Source excerpt

Boltz PBC, a startup that develops artificial intelligence models for pharmaceutical researchers, has raised $110 million in funding. Bessemer led the Series A investment. Boltz stated on Wednesday that the fund was joined by a large group of new and returning backers. The consortium included Greycroft, Andreessen Horowitz, Amplify Partners, F-Prime Capital, Hummingbird and others. [...] The post Medical AI startup Boltz raises $110M to speed up small molecule research appeared first on SiliconANGLE.

## Openhagen Worlds: Calibrated Environments for Measuring and Training Scientific Discovery Agents

DevFeed: [Openhagen Worlds: Calibrated Environments for Measuring and Training Scientific Discovery Agents](<https://devfeed.tech/articles/openhagen-worlds-calibrated-environments-for-measuring-and-training-scientific-discovery-agents-142825.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24711>)

Author: Alexander Boesgaard Lorup

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [budget](<https://devfeed.tech/tags/budget.md>), [chemistry](<https://devfeed.tech/tags/chemistry.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deepseek](<https://devfeed.tech/tags/deepseek.md>), [deepseek-v4-1-flash](<https://devfeed.tech/tags/deepseek-v4-1-flash.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [discovery-agents](<https://devfeed.tech/tags/discovery-agents.md>), [environments](<https://devfeed.tech/tags/environments.md>), [experimental](<https://devfeed.tech/tags/experimental.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [frontier](<https://devfeed.tech/tags/frontier.md>), [iteration](<https://devfeed.tech/tags/iteration.md>), [measuring](<https://devfeed.tech/tags/measuring.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [training](<https://devfeed.tech/tags/training.md>), [worlds](<https://devfeed.tech/tags/worlds.md>)

### AI overview

Openhagen Worlds evaluates scientific discovery agents across 27 procedurally generated environments: agents run budgeted experiments, then commit answers checked against hidden truths by a deterministic verifier. Its 178 difficulty levels are calibrated on 64 development instances using success thresholds for a scripted oracle, plausible non-deductive policies, and gated shortcuts. The 92 core levels cover all families and target small-model training; 86 frontier rungs extend 15 discovery worlds, including multi-stage campaigns. On 84 withheld frontier instances, DeepSeek V4.1 Flash solved 32 at default effort and 57 at maximum effort, versus 79 for the oracle; the largest remaining gap was in eight-stage campaigns. Expert iteration raised a 9B model's core success on trained families from 0.090 to 0.148, partly through improved action formatting, with no detectable GPQA Diamond loss.

### Source excerpt

arXiv:2611.24711v1 Announce Type: cross Abstract: Agents that do science must choose informative experiments and reason from their results, but most discovery benchmarks cannot separate that skill from recall, unstated scoring rules or shortcut policies. We introduce Openhagen Worlds, a suite of 27 procedurally generated environments (some with finite truth spaces), from classic puzzles of experimental reasoning to simulator-grounded physics and chemistry, in which an agent runs experiments under a budget and commits an answer that a deterministic verifier checks against a hidden truth. Every observation states the complete scoring rule, and each of the 178 difficulty levels is calibrated: it is admitted only if, on 64 development instances, a scripted oracle succeeds on at least 0.75, the strongest plausible non-deductive policy on at most 0.25 and each gated shortcut on at most 0.35. The 92 core levels span all 27 families and are sized to train small models; the 86 frontier rungs extend the 15 discovery worlds with harder questions and with campaigns of three, five and eight or more committed stages on one hidden system. On the same 84 withheld frontier instances, one per family and rung, DeepSeek V4.1 Flash solves 32 at its default reasoning effort with 16,384 tokens per call and 57 at its maximum effort with up to 393,216 tokens per call (oracle: 79); the headroom that remains is largest in eight-stage campaigns. On-policy expert iteration gives a 9B model modest gains: its core-tier success on trained families rises from 0.090 to 0.148, partly through better action format, with no detectable loss on GPQA Diamond. A hosted service plays the core tier's withheld instances.

## Tesorai raises $5.3M to feed starving AI models with biological data and accelerate drug discovery

DevFeed: [Tesorai raises $5.3M to feed starving AI models with biological data and accelerate drug discovery](<https://devfeed.tech/articles/tesorai-raises-5-3m-to-feed-starving-ai-models-with-biological-data-and-accelerate-drug-discovery-140710.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/18/tesorai-raises-5-3m-to-feed-starving-ai-models-with-biological-data-and-accelerate-drug-discovery/>)

Author: Mike Wheatley

Published: 2026-11-18T14:00:33Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Bioinformatics](<https://devfeed.tech/topics/bioinformatics.md>), [AI Models](<https://devfeed.tech/topics/ai-models.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-drug-discovery](<https://devfeed.tech/tags/ai-drug-discovery.md>), [ai-insights](<https://devfeed.tech/tags/ai-insights.md>), [ai-model](<https://devfeed.tech/tags/ai-model.md>), [ai-model-training](<https://devfeed.tech/tags/ai-model-training.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [ai-research](<https://devfeed.tech/tags/ai-research.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [big-data](<https://devfeed.tech/tags/big-data.md>), [biological-data](<https://devfeed.tech/tags/biological-data.md>), [clinical-studies](<https://devfeed.tech/tags/clinical-studies.md>), [data](<https://devfeed.tech/tags/data.md>), [drug-discovery](<https://devfeed.tech/tags/drug-discovery.md>), [drug-research](<https://devfeed.tech/tags/drug-research.md>), [funding](<https://devfeed.tech/tags/funding.md>), [genomic-data](<https://devfeed.tech/tags/genomic-data.md>), [melissa-miao](<https://devfeed.tech/tags/melissa-miao.md>), [news](<https://devfeed.tech/tags/news.md>), [peter-cimermancic](<https://devfeed.tech/tags/peter-cimermancic.md>), [protein-data](<https://devfeed.tech/tags/protein-data.md>), [scientific-studies](<https://devfeed.tech/tags/scientific-studies.md>), [seed-funding](<https://devfeed.tech/tags/seed-funding.md>), [startup](<https://devfeed.tech/tags/startup.md>), [tesorai](<https://devfeed.tech/tags/tesorai.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>)

### AI overview

Tesorai raised $5.3 million in seed funding to make existing biological data more useful for AI-driven drug discovery; total funding is now $8 million. Its platform analyzes raw lab-instrument data alongside historical studies: Search ingests and analyzes large datasets, Chat lets scientists query them in plain English, and Trove catalogs thousands of public datasets and studies. The company says reanalysis of five Alzheimer's brain studies surfaced p-tau217, and that adding protein data helped a small single-cell AI model outperform larger RNA-only models on multiple benchmarks. It has partnerships with MD Anderson, NYU and the Michael J. Fox Foundation. Funding will support additional life-sciences partnerships, a larger public-study catalog and coverage of molecular data beyond proteins and genomes.

### Source excerpt

Drug researchers and medical laboratories generate massive volumes of biological data every single day, yet somehow the artificial intelligence models designed to speed up drug discovery are still starved of useful information. It's a strange paradox, but a startup called Tesorai Inc. believes it has the solution after raising $5.3 million in seed funding today. [...] The post Tesorai raises $5.3M to feed starving AI models with biological data and accelerate drug discovery appeared first on SiliconANGLE.

## Colombo: A Frontier-to-Frontier Autoresearch System for Oracle-Based Optimization

DevFeed: [Colombo: A Frontier-to-Frontier Autoresearch System for Oracle-Based Optimization](<https://devfeed.tech/articles/colombo-a-frontier-to-frontier-autoresearch-system-for-oracle-based-optimization-140116.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23080>)

Author: Haihan Zhang, Wendao Wu, Chenheng Zhang, Chunyuan Zheng, Yanyi Li, Cong Fang, Haoxuan Li, Zhouchen Lin

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Combinatorial optimization](<https://devfeed.tech/topics/combinatorial-optimization.md>), [Algorithms](<https://devfeed.tech/topics/algorithms.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Operations research and optimization](<https://devfeed.tech/topics/operations-research-and-optimization.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [adversarial](<https://devfeed.tech/tags/adversarial.md>), [ai](<https://devfeed.tech/tags/ai.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [atlas](<https://devfeed.tech/tags/atlas.md>), [autoresearch](<https://devfeed.tech/tags/autoresearch.md>), [certification](<https://devfeed.tech/tags/certification.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [formal](<https://devfeed.tech/tags/formal.md>), [frontier](<https://devfeed.tech/tags/frontier.md>), [independent](<https://devfeed.tech/tags/independent.md>), [lean](<https://devfeed.tech/tags/lean.md>), [math-oc](<https://devfeed.tech/tags/math-oc.md>), [memory](<https://devfeed.tech/tags/memory.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [oracle](<https://devfeed.tech/tags/oracle.md>), [research](<https://devfeed.tech/tags/research.md>), [science](<https://devfeed.tech/tags/science.md>), [system](<https://devfeed.tech/tags/system.md>)

### AI overview

Colombo is a self-organizing AI research system instantiated in oracle-based optimization. It maps gaps between achievable guarantees and unavoidable costs in a Complexity Atlas, then turns unresolved gaps into adaptive research programmes. Its Adversarial Co-Ascent develops algorithms alongside hard problem families to test information use and expose unavoidable costs. Independent review and Lean-based certification check claims, while long-term memory carries supported findings across settings. The abstract reports optimal or near-optimal rates for more than fifty Atlas settings; it does not provide details of those settings or the underlying proofs.

### Source excerpt

arXiv:2611.23080v1 Announce Type: new Abstract: Science advances from frontier to frontier as discoveries reshape what is known and create the basis for further discovery. We present Colombo, a self-organizing AI researcher that drives this renewal, first instantiated in oracle-based optimization. Here, the central question is how efficiently a problem class can be solved through a prescribed information interface and what costs no admissible algorithm can avoid. The gap between achievable guarantees and unavoidable limits defines the frontier recorded in its Complexity Atlas, a living map of frontier settings and the current state of knowledge. Self-Organization turns unresolved gaps into adaptive research programmes, coordinating conjecture formation, mathematical investigation, and replanning as evidence accumulates. Adversarial Co-Ascent develops algorithms and hard families through reciprocal challenge, exposing mechanisms of information use and concealment that guide stronger methods and proofs of unavoidable costs. Independent review and Lean-based formal certification check mathematical claims, while the Long-Term Memory carries supported findings and research experience across settings. To date, Colombo has established optimal or near-optimal rates for more than fifty settings in the Atlas. Colombo thus takes a concrete step toward Autopoietic Science: a scientific process whose validated advances regenerate the conditions of its own progress, continually reshaping how inquiry is organized and what it can discover.

## SciWorkBench: A Benchmark for Real Scientific Workflows

DevFeed: [SciWorkBench: A Benchmark for Real Scientific Workflows](<https://devfeed.tech/articles/sciworkbench-a-benchmark-for-real-scientific-workflows-140037.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22490>)

Author: Yuchen Zhang, Taolin Han, Jinghang Wang, Jinxin Wang, Yun Wu, Wai Yuet Chiu, Xiaogang Li, Yuting Fu, Kewei Sun, Jiaxin Li, Qile Jin, Jiajia Li, Ruodan Chen, Qingteng Chen, Jialu Nie, Junchao Li, Yifei

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [biology](<https://devfeed.tech/tags/biology.md>), [chemistry](<https://devfeed.tech/tags/chemistry.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [materials-science](<https://devfeed.tech/tags/materials-science.md>), [physics](<https://devfeed.tech/tags/physics.md>), [scientific-research](<https://devfeed.tech/tags/scientific-research.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

SciWorkBench evaluates whether agents can carry out realistic scientific workflows across biology, chemistry, materials science, and physics. Its tasks come from peer-reviewed literature or expert-contributed scenarios and are assessed under a unified execution and grading protocol. The best-performing agent, evaluated within the Claude Code and Codex frameworks, scored a mean of 57.38. The reported weaknesses were translating scientific knowledge into task-specific procedures, integrating evidence across modalities, and scientifically correcting their own work. The authors argue that workflow-based evaluation is important for measuring progress toward reliable AI research systems.

### Source excerpt

arXiv:2611.22490v1 Announce Type: new Abstract: Language models are increasingly being used as agents capable of planning and executing projects, as well as refining their actions through feedback. This shift raises a question: can current agents complete projects like real scientists? To examine this question, we introduce SciWorkBench, a benchmark for evaluating agents' abilities on real science tasks across biology, chemistry, materials science, and physics. SciWorkBench is built from peer-reviewed literature or expert-contributed research scenarios. We also propose a unified scientific analysis framework validated by domain experts, which can categorize most tasks in SciWorkBench. Each task covers one or more stages of this framework. We evaluate up-to-date models within the Claude Code and Codex agent frameworks under a unified execution and grading protocol. The best-performing agent achieves a mean score of 57.38, indicating that current agents still struggle with realistic scientific workflows. Our failure analysis identifies three recurring weaknesses at the workflow level: difficulty translating scientific knowledge into task-specific procedures, weak integration of evidence across different modalities, and limited capacity for scientific self-correction. These findings highlight the importance of workflow-based evaluation for measuring progress toward reliable AI systems for scientific research.

## Dandelion: A Spherical Flower for Neural Simulation of Planetary Dynamics

DevFeed: [Dandelion: A Spherical Flower for Neural Simulation of Planetary Dynamics](<https://devfeed.tech/articles/dandelion-a-spherical-flower-for-neural-simulation-of-planetary-dynamics-141601.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2608.27521>)

Author: Till Muser, Giovanni Abati, Ivan Dokmani\\'c

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>)

Tags: [2d](<https://devfeed.tech/tags/2d.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [all](<https://devfeed.tech/tags/all.md>), [also](<https://devfeed.tech/tags/also.md>), [applying](<https://devfeed.tech/tags/applying.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [attention](<https://devfeed.tech/tags/attention.md>), [based](<https://devfeed.tech/tags/based.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [best](<https://devfeed.tech/tags/best.md>), [but](<https://devfeed.tech/tags/but.md>), [changes](<https://devfeed.tech/tags/changes.md>), [compare](<https://devfeed.tech/tags/compare.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [disco](<https://devfeed.tech/tags/disco.md>), [domain](<https://devfeed.tech/tags/domain.md>), [dry](<https://devfeed.tech/tags/dry.md>), [neural](<https://devfeed.tech/tags/neural.md>), [simulation](<https://devfeed.tech/tags/simulation.md>)

### AI overview

Dandelion adapts the warp-based neural PDE solver Flower to spherical dynamics: each layer predicts a tangent-plane displacement and transports features along great circles, while spherical-harmonic pooling creates a U-Net-like hierarchy. It uses no convolutions; spatial mixing comes from spherical coordinate changes. The authors also release an evolving benchmark suite spanning six spherical PDE settings, intended to support model iteration between small stylized datasets and ERA5-scale data. Dandelion reportedly performs best in all but one benchmark setting, where the warp-based Flower2D leads, and its spherical primitives produce the most physically consistent forecasts.

### Source excerpt

arXiv:2608.27521v2 Announce Type: replace Abstract: Many dynamical processes unfold on the sphere but the default scientific machine learning architectures are Euclidean. Applying these architectures on a regular lat--lon grid causes problems: Cartesian convolutions become distorted at high latitude; 2D FFTs in Fourier neural operators incorrectly assume double periodicity; Cartesian positional encodings in ViTs distort spherical geodesic distances. Recent work moves towards natively spherical primitives, including spherical convolutions (e.g., DeepSphere or DISCO), Spherical Fourier Neural Operators (SFNOs), and geodesic attention. Here we propose Dandelion, a spherical version of Flower---a warp-based neural PDE solver. A Dandelion layer computes a tangent-plane displacement and then transports features along great circles. We obtain a U-Net-like structure by implementing hierarchical pooling entirely in the spherical-harmonic domain. There are thus no convolutions: spatial mixing is achieved only through spherical coordinate changes---or warps. To compare Dandelion with existing spherical architectures, we release an evolving benchmark suite of challenging, natively spherical PDE datasets including a modified Galewsky jet, anomalous chained turbulence, Cahn--Hilliard decomposition, spherical Riemann shocks, Held--Suarez dry atmospheric transport and global ocean dynamics. This new benchmark fills the gap between existing spherical datasets which are either too small and stylized, or much too large (ERA5) for model iteration. Dandelion achieves the best performance on all but one benchmark setting, where the best model is Flower2D, also based on warps. The spherical primitives result in the most physically consistent forecasts.

## Inventa: Decentralized Self-Improving Agents for Autonomous Scientific Discovery

DevFeed: [Inventa: Decentralized Self-Improving Agents for Autonomous Scientific Discovery](<https://devfeed.tech/articles/inventa-decentralized-self-improving-agents-for-autonomous-scientific-discovery-139918.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21847>)

Author: Prithwish Jana, Nikos Kanakaris, Purak Jain, Anoop Deoras, Sahika Genc

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [automated](<https://devfeed.tech/tags/automated.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [decentralized](<https://devfeed.tech/tags/decentralized.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [ideation](<https://devfeed.tech/tags/ideation.md>), [large-language-model](<https://devfeed.tech/tags/large-language-model.md>), [llm](<https://devfeed.tech/tags/llm.md>), [research](<https://devfeed.tech/tags/research.md>), [retrieval](<https://devfeed.tech/tags/retrieval.md>), [self-improving-agents](<https://devfeed.tech/tags/self-improving-agents.md>), [writing](<https://devfeed.tech/tags/writing.md>)

### AI overview

Inventa aims to improve autonomous scientific discovery by combining independent research units with verified shared knowledge. Each unit evolves candidate ideas through a mutator agent; a local orchestrator adapts its setup and, when progress stalls, can replace its model or restart the process. Verifiers record improvements, useful components, and constraint violations in shared knowledge, the only channel between units. Across five benchmarks covering retrieval, ideation, proposal writing, and scientific law discovery, the abstract reports that Inventa leads all five. It scores 73.6% on EurekaBench versus 56.2% for CORAL; on IdeaBench, it finds 26.0 novel ideas per paper versus 18.6 for ResearchAgent, using 82% fewer model calls.

### Source excerpt

arXiv:2611.21847v1 Announce Type: cross Abstract: Large language model (LLM) agents can generate research ideas and improve scientific discovery through iterative experimentation. However, increasing the number of agents or iterations does not automatically translate to producing more diverse or feasible ideas. Agents may converge on similar candidates, propagate unsupported claims through shared memory components and repeatedly apply search strategies that no longer yield any improvements. We introduce Inventa, a framework for automated scientific discovery that employs autonomous research units and verified shared knowledge. Each unit creates a lineage of candidate ideas using a mutator agent. A local orchestrator agent periodically adapts the mutator's harness. When progress stalls, the orchestrator agent replaces its model or restarts the idea evolution process. Verifiers evaluate each candidate idea and record verified improvements, useful components and constraint violations in a shared knowledge, the only channel between units. We evaluate Inventa on five benchmarks spanning the research cycle, covering retrieval (ScholarCatalyst), ideation (IdeaBench), proposal writing (IdeaScientist), and scientific law discovery (EurekaBench, SciLaws-Bench). It leads all five, achieving 73.6% on EurekaBench versus 56.2% for CORAL. On IdeaBench, it discovers 26.0 novel ideas per paper, compared with 18.6 for ResearchAgent, with 82% fewer model calls.

## Alphabet's Isomorphic Labs raises $5B to accelerate drug discovery with AI

DevFeed: [Alphabet's Isomorphic Labs raises $5B to accelerate drug discovery with AI](<https://devfeed.tech/articles/alphabet-s-isomorphic-labs-raises-5b-to-accelerate-drug-discovery-with-ai-139215.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/17/alphabets-isomorphic-labs-raises-5b-to-accelerate-drug-discovery-with-ai/>)

Author: Maria Deutscher

Published: 2026-11-17T23:51:11Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [alphabet](<https://devfeed.tech/tags/alphabet.md>), [biology](<https://devfeed.tech/tags/biology.md>), [clinical-trials](<https://devfeed.tech/tags/clinical-trials.md>), [drug-discovery](<https://devfeed.tech/tags/drug-discovery.md>), [funding-round](<https://devfeed.tech/tags/funding-round.md>), [google](<https://devfeed.tech/tags/google.md>), [investment](<https://devfeed.tech/tags/investment.md>), [isodde](<https://devfeed.tech/tags/isodde.md>), [isomorphic-labs](<https://devfeed.tech/tags/isomorphic-labs.md>), [news](<https://devfeed.tech/tags/news.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>)

### AI overview

Isomorphic Labs closed a $5 billion investment led by Greenoaks and Sequoia Capital, with Alphabet also making a "significant investment" through GV and CapitalG. Its IsoDDE platform is used by multiple pharmaceutical companies, according to the company, to speed drug discovery tasks. The article says IsoDDE helps identify protein binding pockets or develop compounds that induce them. It also says the platform skips the crystallization and X-ray scanning used to create 3D protein models for FEP algorithms, which can be hardware-intensive and costly, and that IsoDDE produces higher-quality output; these are Isomorphic's claims. CEO Demis Hassabis said the funding would support foundational biology models, disease understanding, and therapeutic programs moving toward clinical trials. The company also plans to build bioengineering infrastructure and expand its partner ecosystem.

### Source excerpt

Isomorphic Labs Ltd. today announced that it has closed a $5 billion investment led by Greenoaks and Sequoia Capital. The venture capital firms were joined by more than a half-dozen other backers. The group included Isomorphic parent Alphabet Inc., which made a "significant investment." The search giant provided the capital through its GV and CapitalG startup [...] The post Alphabet's Isomorphic Labs raises $5B to accelerate drug discovery with AI appeared first on SiliconANGLE.

## Autonomous Science Bench: Evaluating AI Agents for Autonomous Scientific Discovery

DevFeed: [Autonomous Science Bench: Evaluating AI Agents for Autonomous Scientific Discovery](<https://devfeed.tech/articles/autonomous-science-bench-evaluating-ai-agents-for-autonomous-scientific-discovery-137294.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.18449>)

Author: Yibo Wen, Han Liu

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Electrical engineering and computer science (EECS)](<https://devfeed.tech/topics/electrical-engineering-and-computer-science-eecs.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [campaign](<https://devfeed.tech/tags/campaign.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [science](<https://devfeed.tech/tags/science.md>)

### AI overview

Autonomous Science Bench evaluates whether AI agents can discover promising designs and learn to rank unseen ones under experimental budget limits. Its initial release includes six campaigns grounded in published biology, chemistry, and materials science experiments: agents submit batches of designs to a lab service, use returned measurements to choose further experiments, then provide a predictive model. Success requires passing separate calibrated gates for measured-design discovery and held-out-design ranking. In 180 pilot trials across ten agent-model configurations, the best succeeded in 72%, while the hardest campaign was solved in five of 30 attempts. The two gates expose distinct failure modes--finding too few promising designs despite accurate rankings, or finding promising designs without predicting unseen performance--that an overall success rate can hide. Version 0.1 includes the campaigns, lab runtime, and public leaderboard.

### Source excerpt

arXiv:2610.18449v2 Announce Type: replace Abstract: AI agents for autonomous scientific discovery must decide which experiments to run under a limited budget. Most benchmarks for science agents provide fixed datasets, while others that support active experimentation largely rely on simulation. We introduce Autonomous Science Bench, an initial release comprising six budgeted design-build-test-learn campaigns grounded in published experiments across biology, chemistry, and materials science. Agents select batches of designs, receive measurements from a lab service, and use the results to guide subsequent experiments before delivering a predictive model. Success requires passing two calibrated gates: a discovery gate, which evaluates the designs the agent measured, and a learning gate, which evaluates its model's ranking of held-out designs. In a pilot evaluation of ten agent-model configurations and 180 trials, the best configuration succeeds in 72% of trials, while the hardest campaign is solved in only five of 30 attempts. The two gates reveal different weaknesses: some agents build accurate rankings but find too few promising designs, while others find promising designs but fail to predict which unseen designs will perform well. These differences are obscured by overall success rates alone. We release version 0.1 of the campaigns, lab runtime, and public leaderboard at https://asb.open-ssi.org/.

## Towards Agentic Scientific Discovery in Distributed Systems

DevFeed: [Towards Agentic Scientific Discovery in Distributed Systems](<https://devfeed.tech/articles/towards-agentic-scientific-discovery-in-distributed-systems-137528.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20031>)

Author: Rashmita Soni, Abhirup Ghosh, Soumya Banerjee

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [distributed](<https://devfeed.tech/tags/distributed.md>)

### AI overview

Query-ATLAS uses an LLM to guide symbolic equation search across clients, drawing on client errors and earlier proposals to adapt tool selection. In the reported comparison with fixed PySR-LLM, results depend on the task: Query-ATLAS improved validation acceptances on CRK3 (7/12 vs. 3/12) and oscillator2 (12/12 vs. 11/12), while MatSci3 favored the fixed workflow. Across 72 settings in a post-hoc six-task subset, Query-ATLAS had 64 validation and 64 in-distribution passes, versus 62 and 59; both workflows passed 47/60 eligible out-of-distribution tests. Fixed PySR-LLM retained lower median errors and cost. These simulated results suggest broader query coverage as a potential benefit, alongside task-dependent tradeoffs.

### Source excerpt

arXiv:2611.20031v1 Announce Type: new Abstract: We investigate when adaptive tool selection helps identify a shared equation across clients with different input coverage. Query-ATLAS (query-conditioned Agentic Tool-based Learning for Aggregating Symbolic Equations) uses an LLM to guide symbolic search from client errors and previous proposals in a simulated distributed system. A four-architecture comparison reveals task-dependent opportunities: relative to fixed PySR-LLM, validation acceptances rise from 3/12 to 7/12 on CRK3 and from 11/12 to 12/12 on oscillator2, while MatSci3 favors the fixed workflow. Across 72 settings in a post-hoc six-task subset of eight development tasks, Query-ATLAS achieves 64 validation and 64 in-distribution passes, versus 62 and 59 for the fixed workflow; both pass 47/60 eligible out-of-distribution tests. The fixed workflow retains lower median errors and cost. Broader query coverage emerges as a promising benefit, motivating task-aware tool selection and client-aware fitting for scientific agents.

## Eliciting Research Taste in LLMs through Future Research Direction Choice

DevFeed: [Eliciting Research Taste in LLMs through Future Research Direction Choice](<https://devfeed.tech/articles/eliciting-research-taste-in-llms-through-future-research-direction-choice-137494.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19794>)

Author: Dhruv Trehan, Paras Chopra

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [Audit-then-score](<https://devfeed.tech/topics/audit-then-score.md>)

Tags: [choice](<https://devfeed.tech/tags/choice.md>), [comparisons](<https://devfeed.tech/tags/comparisons.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [llms](<https://devfeed.tech/tags/llms.md>), [ml](<https://devfeed.tech/tags/ml.md>), [research](<https://devfeed.tech/tags/research.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [steering](<https://devfeed.tech/tags/steering.md>)

### AI overview

The study operationalizes LLM research taste as choosing between future research directions, using human-authored recommendations as a behavioral reference rather than treating them as objectively correct. Its order-controlled evaluation covers 80 comparisons drawn from 19 recent ML papers across four subdomains. Across eight LLMs, author alignment ranges from 45.0% to 67.5%, and stable divergence from 15.0% to 36.2%; divergences largely persist under similarity controls, prompt ablations, and an author-preference framing check. The authors report shared non-author preferences across models, with recurring movement toward scaling or generalization and trade-offs between methodological and infrastructure or data extensions. They present this as an early descriptive evaluation, not a measure of objectively correct research choices.

### Source excerpt

arXiv:2611.19794v1 Announce Type: cross Abstract: Research taste is increasingly discussed in the context of AI scientists, yet remains little studied or measured in LLMs. This judgment matters especially in open-ended research, where a direction must be chosen before a verifiable objective is available. We operationalize research taste through one task of the scientific workflow: choosing which future research direction to pursue given a research goal and current results. Rather than predicting downstream scientometric metrics or treating author choices as objectively correct, we evaluate models against human-authored future-work directions and report author alignment, stable divergence, and unstable choice as behavioral measures of research-direction preference. We construct an order-controlled evaluation of 80 pairwise comparisons from 19 ML papers published between December 2025 and February 2026 across four subdomains. Each query presents a research goal, hypotheses tested with key results, and two candidate future directions: one author-recommended direction and one plausible distractor sampled from a similar paper. Across eight LLMs, author alignment ranges from 45.0% to 67.5%, while stable divergence ranges from 15.0% to 36.2%. These divergences largely persist under similarity controls, prompt ablations, and an author-preference framing check, and inter-model agreement suggests that some non-author preferences are shared across models. We characterize what models diverge toward using transition-category labels and selected reasoning examples, finding recurring movement toward scaling/generalization and trade-offs between methodological and infrastructure/data extensions. We view this as an early descriptive evaluation for eliciting research taste profiles in LLMs, and as a step toward understanding and steering their research preferences as they take on more autonomous roles in novel open-ended scientific discovery.

## Asking the right question is key to AI discovery in a chain of hypotheses

DevFeed: [Asking the right question is key to AI discovery in a chain of hypotheses](<https://devfeed.tech/articles/asking-the-right-question-is-key-to-ai-discovery-in-a-chain-of-hypotheses-137501.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19841>)

Author: Hu Cang

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Occam's razor machine learning](<https://devfeed.tech/topics/occam-s-razor-machine-learning.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [all](<https://devfeed.tech/tags/all.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [artificial-intelligence-ai](<https://devfeed.tech/tags/artificial-intelligence-ai.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [candidate](<https://devfeed.tech/tags/candidate.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-sonnet](<https://devfeed.tech/tags/claude-sonnet.md>), [claude-sonnet-5-5](<https://devfeed.tech/tags/claude-sonnet-5-5.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [data](<https://devfeed.tech/tags/data.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [existing](<https://devfeed.tech/tags/existing.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [feature](<https://devfeed.tech/tags/feature.md>), [frontier](<https://devfeed.tech/tags/frontier.md>), [key](<https://devfeed.tech/tags/key.md>), [physics-optics](<https://devfeed.tech/tags/physics-optics.md>), [question](<https://devfeed.tech/tags/question.md>), [right](<https://devfeed.tech/tags/right.md>)

### AI overview

Across 32 closed-loop trials with a narrow peak missing from the initial data, Claude Sonnet 5.5 and GPT-6.1-Sol sampled within the peak region less often than random search. The agent that followed model disagreement more closely missed the peak in every trial. The abstract attributes this failure to candidate models agreeing across an unmeasured gap, leaving disagreement-guided search little reason to probe it. Discovery succeeded when experiments tested the gap early and then followed up across the anomaly. The findings suggest that, in this setup, probing plausible unmeasured regions can matter more than choosing experiments solely by model disagreement.

### Source excerpt

arXiv:2611.19841v1 Announce Type: new Abstract: Artificial intelligence (AI) scientists pursue discovery through a chain of hypotheses, fitting candidate models to existing observations and often choosing experiments where those models disagree most. Yet when early data miss a narrow feature, model selection favors simple curves that agree across the unmea- sured gap, so disagreement-guided search has no reason to look where the new phenomenon lies. When we tested two frontier AI scientists, Claude Sonnet 5.5 and GPT-6.1-Sol, across 32 closed-loop trials with a narrow peak omitted from the starting data, both sampled inside the peak region less often than random search. Surprisingly, the agent that followed model disagreement more closely missed the peak in all 32 trials, whereas discovery succeeded when experiments probed the unmeasured gap early and followed up across the anomaly. Asking the right question, rather than testing only where current models disagree, is therefore key to AI discovery in a chain of hypotheses.

## Debate as Reward: A Multi-Agent Reward System for Scientific Ideation via RL Post-Training

DevFeed: [Debate as Reward: A Multi-Agent Reward System for Scientific Ideation via RL Post-Training](<https://devfeed.tech/articles/debate-as-reward-a-multi-agent-reward-system-for-scientific-ideation-via-rl-post-training-137247.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2604.16723>)

Author: Moein Salimi, Babak Hosseini Mohtasham, Amin Aghakasiri, Mahdi Naieni, Amir Hossein Qeysarbeigi, Mohammad Masih Shalchian Nazer, Zahra Azar, Mahdi Jafari Siavoshani, Mohammad Hossein Rohban

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [post-training](<https://devfeed.tech/topics/post-training.md>), [grpo](<https://devfeed.tech/topics/grpo.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>)

Tags: [bias](<https://devfeed.tech/tags/bias.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [effectiveness](<https://devfeed.tech/tags/effectiveness.md>), [group-relative-policy-optimization](<https://devfeed.tech/tags/group-relative-policy-optimization.md>), [iclr](<https://devfeed.tech/tags/iclr.md>), [ideation](<https://devfeed.tech/tags/ideation.md>), [large-language-models-llms](<https://devfeed.tech/tags/large-language-models-llms.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [neurips](<https://devfeed.tech/tags/neurips.md>), [post-training](<https://devfeed.tech/tags/post-training.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [reward](<https://devfeed.tech/tags/reward.md>), [rl](<https://devfeed.tech/tags/rl.md>)

### AI overview

The paper proposes training language models for scientific ideation with a multi-agent judge that separates methodological validation from implementation details and awards strict binary rewards, aiming to reduce reward hacking. To optimize this sparse signal, it uses an unbiased variant of Group Relative Policy Optimization intended to mitigate artificial length bias. Training uses problem-solution pairs from ICLR 2024 and NeurIPS 2025 proceedings. The abstract reports that the framework outperforms state-of-the-art baselines on expert-evaluated novelty, feasibility, and effectiveness; it provides no specific scores or experimental details.

### Source excerpt

arXiv:2604.16723v2 Announce Type: replace Abstract: Large Language Models (LLMs) have demonstrated potential in automating scientific ideation, yet current approaches relying on iterative prompting or complex multi-agent architectures often suffer from hallucination or computational inefficiency. A critical bottleneck in applying Reinforcement Learning (RL) to this open-ended domain is reward hacking---where models exploit imperfect evaluation proxies to maximize scores without producing genuine scientific innovation. To address these limitations, we propose an RL framework explicitly tailored for high-quality scientific idea generation. We propose the first multi-agent reward function designed to serve as a judge, decoupling methodological validation from implementation details while providing strict binary rewards that are robust to reward hacking. To effectively optimize against this sparse signal, we utilize an unbiased variant of Group Relative Policy Optimization to mitigate artificial length bias. We ground our training in a curated dataset of problem-solution pairs extracted from ICLR 2024 and NeurIPS 2025 proceedings. Experiments demonstrate that our framework significantly outperforms state-of-the-art baselines across expert-evaluated metrics of novelty, feasibility, and effectiveness.

## At SC26, Nvidia debuts new Rubin HPC chip, scientific research libraries

DevFeed: [At SC26, Nvidia debuts new Rubin HPC chip, scientific research libraries](<https://devfeed.tech/articles/at-sc26-nvidia-debuts-new-rubin-hpc-chip-scientific-research-libraries-136387.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/16/nvidia-debuts-new-rubin-hpc-chip-scientific-research-libraries/>)

Author: Maria Deutscher

Published: 2026-11-17T01:45:14Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [CUDA-X](<https://devfeed.tech/topics/cuda-x.md>), [Nvidia](<https://devfeed.tech/topics/nvidia.md>), [Nemotron](<https://devfeed.tech/topics/nemotron.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [astrophysics](<https://devfeed.tech/tags/astrophysics.md>), [biology](<https://devfeed.tech/tags/biology.md>), [cuda-x](<https://devfeed.tech/tags/cuda-x.md>), [cuphoton](<https://devfeed.tech/tags/cuphoton.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [libraries](<https://devfeed.tech/tags/libraries.md>), [llm](<https://devfeed.tech/tags/llm.md>), [nasdaq-nvda](<https://devfeed.tech/tags/nasdaq-nvda.md>), [nemotron-3-5-super-vl](<https://devfeed.tech/tags/nemotron-3-5-super-vl.md>), [news](<https://devfeed.tech/tags/news.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvlink](<https://devfeed.tech/tags/nvlink.md>), [rubin](<https://devfeed.tech/tags/rubin.md>), [rubin-hpc](<https://devfeed.tech/tags/rubin-hpc.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [scientific-research](<https://devfeed.tech/tags/scientific-research.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [top-story-5](<https://devfeed.tech/tags/top-story-5.md>), [top-story-6](<https://devfeed.tech/tags/top-story-6.md>)

### AI overview

Nvidia announced a science-focused hardware and software lineup at SC26. Its planned Rubin HPC GPU, due in 2028, will have more than 50,000 CUDA cores, a new NVLink iteration, and sparse-data acceleration; the current Rubin GPU has over 28,000 CUDA cores. New CUDA-X tools include cuPhoton.xBeam for Event Horizon Telescope analysis, which Nvidia says can reduce tasks from a month to eight minutes, cuPhoton.xLens for fusion-plasma simulation, and open-source cuSpa for cell-polygon calculations. Nvidia also packaged its Nemotron 3.5 Super VL model with scientific post-training workflows and nearly 50,000 chemistry questions. CUDA-Q additions include quantum software components and an error-mitigation algorithm; Nvidia, IBM, and Fermilab are also developing a platform combining DGX B200 systems with IBM quantum hardware.

### Source excerpt

Nvidia Corp. today introduced a set of offerings designed to speed up scientific research across a range of fields. The lineup is headlined by Rubin HPC, an upcoming graphics card with significantly more cores than the company's current flagship chip. It's joined by more than a half-dozen software tools geared toward scientists and engineers. The [...] The post At SC26, Nvidia debuts new Rubin HPC chip, scientific research libraries appeared first on SiliconANGLE.

## NVIDIA CUDA-X Libraries Give Agents the Tools to Accelerate Scientific Discovery

DevFeed: [NVIDIA CUDA-X Libraries Give Agents the Tools to Accelerate Scientific Discovery](<https://devfeed.tech/articles/nvidia-cuda-x-libraries-give-agents-the-tools-to-accelerate-scientific-discovery-136621.md>)

Original publisher: [Read original article](<https://blogs.nvidia.com/blog/cuda-x-libraries-agents-accelerate-scientific-discovery/>)

Author: Kibibi Moseley

Published: 2026-11-16T16:00:16Z

Content type: article

Language: en

Sources: [NVIDIA](<https://devfeed.tech/sources/nvidia-blog.md>)

Topics: [CUDA-X](<https://devfeed.tech/topics/cuda-x.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Nvidia](<https://devfeed.tech/topics/nvidia.md>)

Tags: [accelerate](<https://devfeed.tech/tags/accelerate.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [alchemi](<https://devfeed.tech/tags/alchemi.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [astrophysics](<https://devfeed.tech/tags/astrophysics.md>), [biology](<https://devfeed.tech/tags/biology.md>), [complex-systems](<https://devfeed.tech/tags/complex-systems.md>), [cpus](<https://devfeed.tech/tags/cpus.md>), [cuda-x](<https://devfeed.tech/tags/cuda-x.md>), [cuphoton](<https://devfeed.tech/tags/cuphoton.md>), [data](<https://devfeed.tech/tags/data.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [dgx-spark](<https://devfeed.tech/tags/dgx-spark.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [drug-discovery](<https://devfeed.tech/tags/drug-discovery.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [materials](<https://devfeed.tech/tags/materials.md>), [materials-science](<https://devfeed.tech/tags/materials-science.md>), [nemotron](<https://devfeed.tech/tags/nemotron.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [science](<https://devfeed.tech/tags/science.md>), [single-cell](<https://devfeed.tech/tags/single-cell.md>), [software](<https://devfeed.tech/tags/software.md>), [supercomputing](<https://devfeed.tech/tags/supercomputing.md>)

### AI overview

Expanded NVIDIA CUDA-X libraries accelerate scientific data analysis across several fields. cuPhoton.xBeam reduced Event Horizon Telescope correlation from over a month to eight minutes; cuPhoton.xLens can extract insights during plasma simulations, with a locally running Nemotron-powered agent supporting interactive analysis. In molecular simulation, cuPotential reports up to 3x acceleration alongside improved stability and accuracy. Other examples include Daviator's quantum calculations at up to 22x faster and 75% lower cost, cuSpa's 900x speedup for cell polygon calculations, and rapids-singlecell processing reported 200x faster at Therna Biosciences. GPU-accelerated HMMER searches at Basecamp Research ran up to 84x faster than standard HMMER. These reported gains vary by tool and task; the excerpt does not provide a common benchmark for comparing them.

### Source excerpt

With expanded NVIDIA CUDA-X libraries, including cuPhoton, cuPotential and cuSpa, researchers can sift through vast amounts of data to accelerate research, from astrophysics and materials science to single-cell biology. The speedups are helping scientists analyze larger datasets and model more complex systems. The new libraries give AI agents the tools to help researchers make discoveries [...]

## NVIDIA Announces Rubin HPC -- A New GPU to Accelerate Agents for Science, Engineering and Analytics

DevFeed: [NVIDIA Announces Rubin HPC -- A New GPU to Accelerate Agents for Science, Engineering and Analytics](<https://devfeed.tech/articles/nvidia-announces-rubin-hpc-a-new-gpu-to-accelerate-agents-for-science-engineering-and-analytics-136625.md>)

Original publisher: [Read original article](<https://blogs.nvidia.com/blog/rubin-hpc/>)

Author: Dan Ernst

Published: 2026-11-16T16:00:01Z

Content type: release

Language: en

Sources: [NVIDIA](<https://devfeed.tech/sources/nvidia-blog.md>)

Topics: [GPU](<https://devfeed.tech/topics/gpu.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [CUDA-X](<https://devfeed.tech/topics/cuda-x.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [NVIDIA Vera Rubin](<https://devfeed.tech/topics/nvidia-vera-rubin.md>), [Simulation and Design](<https://devfeed.tech/topics/simulation-and-design.md>)

Tags: [accelerate](<https://devfeed.tech/tags/accelerate.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [cuda-x](<https://devfeed.tech/tags/cuda-x.md>), [data](<https://devfeed.tech/tags/data.md>), [data-processing](<https://devfeed.tech/tags/data-processing.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvidia-rubin](<https://devfeed.tech/tags/nvidia-rubin.md>), [nvidia-vera-rubin](<https://devfeed.tech/tags/nvidia-vera-rubin.md>), [nvlink](<https://devfeed.tech/tags/nvlink.md>), [rubin](<https://devfeed.tech/tags/rubin.md>), [rubin-hpc](<https://devfeed.tech/tags/rubin-hpc.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [science](<https://devfeed.tech/tags/science.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [simulations](<https://devfeed.tech/tags/simulations.md>), [supercomputing](<https://devfeed.tech/tags/supercomputing.md>), [test](<https://devfeed.tech/tags/test.md>), [vera-rubin](<https://devfeed.tech/tags/vera-rubin.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

NVIDIA says Rubin HPC, a GPU for the Vera Rubin platform announced at SC26, is designed to speed simulations and data processing used in science, engineering and analytics. It combines more than 50,000 CUDA cores with sparse data acceleration to keep cores supplied with data, plus seventh-generation NVLink for GPU-to-GPU connectivity. The design targets recurring workloads such as sparse matrix operations, stencil computations, sorting and reductions, balancing data movement with compute. NVIDIA CUDA-X libraries provide optimized algorithms intended to work with the GPU. Faster simulations could give agents and researchers quicker feedback as they test and refine candidate ideas. Rubin HPC is expected to be available in 2028.

### Source excerpt

Announced today at SC26, NVIDIA Rubin HPC is a new GPU in the NVIDIA Vera Rubin platform designed to accelerate simulation and data processing for science, engineering and analytics workflows. As agents help scientists and engineers explore more candidate materials, molecules and designs, they rely on simulations to test those ideas and guide their next [...]

## Post-Training NVIDIA Nemotron to Increase Accuracy in Your Scientific Domain

DevFeed: [Post-Training NVIDIA Nemotron to Increase Accuracy in Your Scientific Domain](<https://devfeed.tech/articles/post-training-nvidia-nemotron-to-increase-accuracy-in-your-scientific-domain-136679.md>)

Original publisher: [Read original article](<https://developer.nvidia.com/blog/post-training-nvidia-nemotron-to-increase-accuracy-in-your-scientific-domain/>)

Author: Gautham Dharuman

Published: 2026-11-16T16:00:00Z

Content type: tutorial

Language: en

Sources: [NVIDIA Developer](<https://devfeed.tech/sources/nvidia-developer.md>), [NVIDIA Technical Blog](<https://devfeed.tech/sources/nvidia-technical-blog.md>)

Topics: [post-training](<https://devfeed.tech/topics/post-training.md>), [Nemotron](<https://devfeed.tech/topics/nemotron.md>), [rlvr](<https://devfeed.tech/topics/rlvr.md>), [grpo](<https://devfeed.tech/topics/grpo.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [NeMo](<https://devfeed.tech/topics/nemo.md>)

Tags: [17](<https://devfeed.tech/tags/17.md>), [adapter](<https://devfeed.tech/tags/adapter.md>), [adapters](<https://devfeed.tech/tags/adapters.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agentic-ai-generative-ai](<https://devfeed.tech/tags/agentic-ai-generative-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [autonomous-agents](<https://devfeed.tech/tags/autonomous-agents.md>), [computational-chemistry-materials-science](<https://devfeed.tech/tags/computational-chemistry-materials-science.md>), [developer-tools-techniques](<https://devfeed.tech/tags/developer-tools-techniques.md>), [featured](<https://devfeed.tech/tags/featured.md>), [innovation](<https://devfeed.tech/tags/innovation.md>), [machine-learning-artificial-intelligence](<https://devfeed.tech/tags/machine-learning-artificial-intelligence.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [nemo](<https://devfeed.tech/tags/nemo.md>), [nemotron](<https://devfeed.tech/tags/nemotron.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvidia-nemotron](<https://devfeed.tech/tags/nvidia-nemotron.md>), [post-training](<https://devfeed.tech/tags/post-training.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [research](<https://devfeed.tech/tags/research.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [scientific-computing](<https://devfeed.tech/tags/scientific-computing.md>), [scientific-research](<https://devfeed.tech/tags/scientific-research.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

The excerpt describes adapting Nemotron 3.5 Super VL to scientific tasks with reinforcement learning from verifiable rewards. Experimental records become prompts paired with reference outcomes; a task-specific verifier scores model submissions, so the reward comes from data without human preference labels. In the drug-metabolism example, Morgan fingerprints and Tanimoto similarity score predicted metabolites. The recipe uses NeMo Gym for the environment and NeMo RL for GRPO training with LoRA. It reports SFT improved accuracy by 12%, with RL adding 17% at a 16,384-token output budget. An optional warm start distills teacher traces that meet a similarity threshold; the excerpt reports retaining 1,123 of 1,934 prompts at a threshold of at least 0.7. It emphasizes choosing a reward that reflects the scientific outcome and evaluating on held-out data. These figures and implementation details are specific to the described example.

### Source excerpt

Agentic AI is reshaping scientific research and innovation. Researchers communicate complex scientific hypotheses in natural language and let autonomous agents...

## Science at the Frontier: How scientists use ChatGPT

DevFeed: [Science at the Frontier: How scientists use ChatGPT](<https://devfeed.tech/articles/science-at-the-frontier-how-scientists-use-chatgpt-136431.md>)

Original publisher: [Read original article](<https://openai.com/index/science-at-the-frontier>)

Published: 2026-11-16T12:00:00Z

Content type: article

Language: en

Sources: [OpenAI News](<https://devfeed.tech/sources/openai-news.md>)

Topics: [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [ChatGPT](<https://devfeed.tech/topics/chatgpt.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [academic](<https://devfeed.tech/tags/academic.md>), [accelerate](<https://devfeed.tech/tags/accelerate.md>), [agi](<https://devfeed.tech/tags/agi.md>), [ai](<https://devfeed.tech/tags/ai.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [analyze](<https://devfeed.tech/tags/analyze.md>), [automated](<https://devfeed.tech/tags/automated.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [chatgpt](<https://devfeed.tech/tags/chatgpt.md>), [classifiers](<https://devfeed.tech/tags/classifiers.md>), [code](<https://devfeed.tech/tags/code.md>), [data](<https://devfeed.tech/tags/data.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [explore](<https://devfeed.tech/tags/explore.md>), [frontier](<https://devfeed.tech/tags/frontier.md>), [productivity](<https://devfeed.tech/tags/productivity.md>), [research](<https://devfeed.tech/tags/research.md>), [science](<https://devfeed.tech/tags/science.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

An analysis of 580,186 messages from 1,819 scientists in a ChatGPT academic research pilot found use across the research process, including ideation, evidence analysis, coding, and communicating findings. In messages about research ideas, 35% started with a scientist's specific question, 36% developed a scientist's idea with ChatGPT, and 23% asked ChatGPT to suggest a question. Surveyed participants estimated that equivalent work at the same quality would take 1.9 times as long without AI; 98% of those answering expected AI to accelerate discovery in their field. Among 367 reasons in written responses for expecting faster discovery, 52% cited getting research done faster, 23% finding more promising questions, and 14% pursuing a wider range of questions. These are reported use patterns and expectations, not evidence that AI has already accelerated discoveries. The article says more research is needed and notes that testing ideas also depends on funding, staff, and experimental capacity. Researchers' messages and responses were not read by the study team; automated classifiers categorized messages by activity.

### Source excerpt

Explore how more than 1,800 scientists use ChatGPT in research. See findings on daily workflows, reported productivity, and expectations for discovery.

## Generative machine learning for protein design

DevFeed: [Generative machine learning for protein design](<https://devfeed.tech/articles/generative-machine-learning-for-protein-design-134970.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.26378>)

Author: Senura Hansaja Wanasekara, Manoj M Wagle, Minh-Duong Nguyen, Xiaochen Liu, Nguyen H. Tran, Manolis Kellis, Ken-Tye Yong

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Model Development](<https://devfeed.tech/topics/model-development.md>)

Tags: [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [generative](<https://devfeed.tech/tags/generative.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [protein-design](<https://devfeed.tech/tags/protein-design.md>), [proteins](<https://devfeed.tech/tags/proteins.md>)

### AI overview

The review surveys 59 generative approaches across six protein design tasks: sequence and backbone generation, structure-conditioned sequence and joint sequence-structure design, motif scaffolding, and binder and antibody design. It covers protein representations, how models use structural constraints and functional information, and computational and experimental evaluation. The abstract identifies two limitations: designing proteins outside well-represented families and incorporating experimental feedback. It does not provide comparative results or specific recommendations.

### Source excerpt

arXiv:2603.26378v2 Announce Type: replace-cross Abstract: Designing proteins with specific biological functions is of considerable interest for both the development of new therapeutics and the engineering of enzymes, yet achieving the intended activity remains challenging. Generative artificial intelligence (AI) has emerged as a powerful tool to explore the vast space of possible designs by learning from protein sequence and structural data. Here, we review 59 generative approaches across six protein design tasks, including sequence generation, backbone generation, structure-conditioned sequence design, joint sequence-structure design, motif scaffolding, and binder and antibody design. We outline common generative approaches and protein representations and discuss how structural constraints and functional information are utilized during generation. This is followed by a discussion of the computational and experimental approaches used to evaluate designed proteins and what such assessments reveal about their properties. Finally, we discuss current limitations in designing proteins beyond well-represented families and incorporating experimental feedback into the design process. We expect that rapid advances in generative modeling will enable more reliable design of a wider range of protein structures and functions over the coming years.

## NeoResearch: Literature- and Experience-Guided Autonomous Time Series Model Discovery

DevFeed: [NeoResearch: Literature- and Experience-Guided Autonomous Time Series Model Discovery](<https://devfeed.tech/articles/neoresearch-literature-and-experience-guided-autonomous-time-series-model-discovery-135880.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18235>)

Author: Ze Guo, Mingyue Cheng, Xiaoyu Tao, Shuo Yu, Shangtong Ye

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [conditional](<https://devfeed.tech/tags/conditional.md>), [context](<https://devfeed.tech/tags/context.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [experience](<https://devfeed.tech/tags/experience.md>), [experimentation](<https://devfeed.tech/tags/experimentation.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [exploration](<https://devfeed.tech/tags/exploration.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [framework](<https://devfeed.tech/tags/framework.md>), [grounded](<https://devfeed.tech/tags/grounded.md>), [literature](<https://devfeed.tech/tags/literature.md>), [model](<https://devfeed.tech/tags/model.md>), [series](<https://devfeed.tech/tags/series.md>), [time](<https://devfeed.tech/tags/time.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

NeoResearch uses large language models to discover lightweight forecasting models tailored to time-series tasks. Its dual-loop workflow combines hypothesis-driven architecture exploration with iterative training and hyperparameter optimization guided by experimental feedback. The agent mines web literature for modeling mechanisms, empirical evidence, and applicability conditions, then turns those findings into testable, task-grounded hypotheses. A hierarchical experience memory connects literature findings with experiment results and stores conditional lessons for reuse across tasks. Experiments on ten datasets reportedly show competitive performance and support the value of the dual loops, literature extraction, and memory reuse; the abstract provides no specific baselines or scores.

### Source excerpt

arXiv:2611.18235v1 Announce Type: new Abstract: Developing effective forecasting models for diverse time-series domains requires substantial expert knowledge, evidence from scientific literature on the web, and iterative experimentation. However, translating this knowledge and research experience into model designs requires preserving source context and testing applicability under task-specific assumptions. In this paper, we propose NeoResearch, a literature- and experience-guided autonomous research framework that leverages large language models (LLMs) as researchers to discover lightweight, domain-specific forecasting models. Specifically, NeoResearch establishes an expert-inspired dual-loop research workflow, coupling hypothesis-driven architecture exploration with iterative training and hyperparameter optimization through experimental feedback. To enable knowledge-grounded discovery, the agent autonomously mines scientific literature from the Web, extracting modeling mechanisms, empirical evidence, and applicability conditions to formulate task-grounded, testable architecture hypotheses. Furthermore, a hierarchical experience memory links literature findings to experimental outcomes and consolidates conditional lessons, guiding architecture discovery and training optimization while supporting reuse across forecasting tasks. Experiments on ten datasets show competitive performance and support dual loops, literature extraction, and memory reuse. Our code is available.

[Next page](<https://devfeed.tech/topics/ai-for-science.md?cursor=WyIyMDI2LTExLTE2VDA1OjAwOjAwKzAwOjAwIiwgIjJkMTBjYzJiLWQ4MzEtNDI3NC1iMmI3LTgwMGRlNTkzNDM1MSJd>)