# Amazon Science homepage Learn about Amazon's scientific research, science community, and career opportunities in artificial intelligence (AI), machine learning (ML), computer vision, robotics, quantum, economics and more. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Why don't machine learning research agents overfit? DevFeed: [Why don't machine learning research agents overfit?]() Original publisher: [Read original article]() Author: Martin Bertran Lopez; Aaron Roth Published: 2026-09-10T15:03:39Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [machine learning overfitting](), [Training AI Models](), [AI research agents](), [Occam's razor machine learning]() Tags: [agentic-ai](), [agents](), [ai-research-agents](), [benchmark](), [benchmark-overfitting-machine-learning](), [compressibility-and-memorization](), [compression-and-generalization](), [generalization-in-machine-learning](), [generative-ai](), [information-bottleneck-overfitting](), [language-models](), [llm-compression-theory](), [machine-learning](), [machine-learning-overfitting](), [machine-learning-research](), [occam-s-razor-machine-learning](), [research](), [validation](), [why-don-t-ml-models-overfit-on-benchmarks]() ### AI overview The article explains why repeated evaluation on held-out benchmarks can cause overfitting, then frames the apparent contradiction in machine learning research, where benchmark-driven iteration is widespread. It also summarizes research suggesting that compressible models limit memorization. ### Source excerpt New research indicates that AI agents learn compressible models of data, which don't have enough space to enable memorization. ## Developing provably correct Rust code with Verus DevFeed: [Developing provably correct Rust code with Verus]() Original publisher: [Read original article]() Author: Bryan Parno Published: 2026-08-31T15:35:33Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Rust formal verification](), [Vulnerabilities]() Tags: [amazon-elastic-compute](), [automated-reasoning](), [aws-lambda](), [code](), [firecracker](), [formal-methods](), [formal-verification](), [open-source](), [rust](), [security](), [testing]() ### AI overview The article explains how Verus formally verifies Rust code against mathematical specifications, helping establish correctness beyond what Rust safety checks and conventional tests provide. ### Source excerpt How the Verus "program verifier", which automatically checks code against a mathematical specification of its functionality, helps increase security assurance in software projects. ## When LLM judges agree, should we believe them? DevFeed: [When LLM judges agree, should we believe them?]() Original publisher: [Read original article]() Author: Krishna Balasubramanian; Sasha Podkopaev Published: 2026-08-26T17:10:40Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Large Language Model](), [Retrieval-Augmented Generation](), [Ising](), [benchmark overfitting machine learning](), [Network](), [AI, ML & Data Engineering]() Tags: [conversational-ai](), [evaluation](), [ising](), [large-language-models-llms](), [llm](), [machine-learning](), [metrics](), [retrieval](), [retrieval-augmented-generation]() ### AI overview The article examines whether agreement among LLM judges is trustworthy when their outputs may be correlated. It presents a dependence-aware aggregation method based on Ising models that discounts shared blind spots and outperforms historical-accuracy-weighted majority voting on three tasks. ### Source excerpt Discounting the opinions of LLM judges with highly correlated outputs ensures that panels of judges reflect a true diversity of perspectives. ## SOP-Bench: A new benchmark for evaluating AI agents on real business procedures DevFeed: [SOP-Bench: A new benchmark for evaluating AI agents on real business procedures]() Original publisher: [Read original article]() Author: Rohith Nama; Nandi Subhrangshu Published: 2026-08-21T15:57:17Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Benchmark](), [Artificial Intelligence](), [Ground truth / benchmark quality]() Tags: [agentic-ai](), [ai](), [ai-agents](), [benchmark](), [benchmarks](), [conversational-ai](), [testing]() ### AI overview SOP-Bench is an openly available benchmark for evaluating how well AI agents execute real standard operating procedures authored by domain experts. It combines genuine enterprise procedures, functioning tools, and ground-truth answers to test interpretation, memory, judgment, and tool selection during complete procedures. ### Source excerpt Extendable framework enables testing agents on the full set of capabilities required to successfully complete a procedure, not isolated proxy tasks. ## A decade of mathematical certainty: Reflections on the Automated Reasoning Group DevFeed: [A decade of mathematical certainty: Reflections on the Automated Reasoning Group]() Original publisher: [Read original article]() Author: Byron Cook Published: 2026-08-11T16:22:19Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Automated reasoning](), [Formal verification](), [Math and Logic](), [systems](), [Amazon Web Services](), [VPC](), [network security](), [Security](), [AWS IAM](), [Amazon S3](), [Cloud](), [amazon]() Tags: [agentic-ai](), [amazon](), [amazon-web-services-aws](), [automated-reasoning](), [aws](), [compliance](), [formal-verification](), [network-security](), [security](), [security-privacy-and-abuse-prevention](), [vpc](), [vulnerabilities]() ### AI overview Amazon's Automated Reasoning Group reflects on a decade of applying mathematical logic, formal verification, and program analysis to AWS security and reliability. The article describes how research projects became production systems, including Tiros for VPC and network analysis and Zelkova for analyzing policies, S3 Block Public Access, and IAM Access Analyzer. ### Source excerpt Ten years after we founded the Automated Reasoning Group, mathematical logic has moved from academic research into production services that secure millions of customer workloads -- demonstrating that systems can be provably correct, not just probably correct. ## AWS Trainium Frontier competition: Co-design models and kernels on purpose-built AI chips DevFeed: [AWS Trainium Frontier competition: Co-design models and kernels on purpose-built AI chips]() Original publisher: [Read original article]() Author: Louise Ping; John Gray; Emily Webber; Josh Longenecker Published: 2026-08-10T20:23:04Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [AI Chat]() Tags: [aws-trainium](), [chip-design](), [hardware](), [kernels](), [language-models](), [machine-learning](), [model-architecture](), [neurips](), [performance](), [training]() ### AI overview AWS Trainium Frontier is a competition for training language models from scratch on Trainium while co-designing architectures, optimizers, training loops, and optional custom kernels under fixed compute and time budgets. ### Source excerpt A competition with a finalist ceremony during NeurIPS 2026, challenging researchers to train language models from scratch on Trainium, exploring what optimal architectures look like when the hardware changes. ## 34 Amazon Research Awards Build on Trainium recipients announced DevFeed: [34 Amazon Research Awards Build on Trainium recipients announced]() Original publisher: [Read original article]() Author: Amazon Research Awards team Published: 2026-08-05T15:00:00Z Content type: news Language: en Sources: [Amazon Science homepage]() Topics: [Artificial Intelligence](), [responsible-ai](), [AWS AI chips](), [AI Research](), [Optimization](), [Inference](), [moe](), [Large Language Model](), [distributed-systems](), [synthetic-data]() Tags: [academic-ai-funding](), [ai](), [ai-research](), [ai-research-grants](), [ai-safety-and-alignment](), [amazon-research-awards](), [ara](), [aws-ai-chips](), [aws-trainium](), [build-on-trainium](), [data](), [deep-learning](), [distributed-systems](), [inference](), [internal-ara-program-updates](), [llm](), [machine-learning-research](), [mixture-of-experts](), [optimization](), [research](), [responsible-ai](), [synthetic](), [synthetic-data-generation]() ### AI overview Amazon announces 34 recipients of its Build on Trainium program, a $110 million credit initiative supporting AI research and university education. The awards fund work in areas including Responsible AI, language models, synthetic data, distributed systems, model architectures, libraries, and optimization on AWS Trainium. ### Source excerpt Amazon announces 34 recipients of the Build on Trainium program, a $110 million credit initiative supporting AI research at 30 universities including Stanford, UC Berkeley, UIUC, UCLA, CMU, and MIT, with a focus on Responsible AI. ## How controllers from industrial machinery can coordinate multitask machine learning DevFeed: [How controllers from industrial machinery can coordinate multitask machine learning]() Original publisher: [Read original article]() Author: Theodore Vasiloudis Published: 2026-07-30T17:26:47Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Machine Learning & Artificial Intelligence](), [Graphs](), [Training AI Models](), [Neural Network](), [data](), [Algorithm](), [systems]() Tags: [data](), [machine-learning](), [model](), [multitask-learning](), [research](), [self-supervised-learning](), [train](), [training]() ### AI overview ControlG addresses conflicting objectives in graph self-supervised learning by allocating computational capacity to one objective at a time and using a proportional-integral-derivative controller to select which objective receives attention next. ### Source excerpt Instead of compromising among parameter updates dictated by different training objectives, ControlG allocates computational capacity to objectives sequentially and dynamically. ## A new benchmark for evaluating patient-facing health AI agents DevFeed: [A new benchmark for evaluating patient-facing health AI agents]() Original publisher: [Read original article]() Author: Korosh Vatanparvar; Ashutosh Joshi Published: 2026-07-29T15:16:52Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [AI Bots]() Tags: [agentic-ai](), [agents](), [ai](), [benchmark](), [conversational-ai](), [evaluation](), [foundation-models](), [health](), [healthcare](), [safety](), [synthetic]() ### AI overview The article introduces PatientAgentBench, a clinician-vetted benchmark for evaluating the safety and task performance of patient-facing healthcare AI agents in realistic, multiturn conversations. ### Source excerpt PatientAgentBench generates a synthetic patient health record, a realistic clinical vignette, and a patient agent that converses with the AI system under evaluation, to capture what a patient-facing agent actually has to do. ## Amazon is investing in the Lean Focused Research Organization DevFeed: [Amazon is investing in the Lean Focused Research Organization]() Original publisher: [Read original article]() Author: Byron Cook; Shawn Bice Published: 2026-07-26T08:00:00Z Content type: news Language: en Sources: [Amazon Science homepage]() Topics: [AI Bots](), [AI Chat]() Tags: [agents](), [ai-agents](), [amazon](), [amazon-aurora](), [amazon-bedrock](), [amazon-bedrock-agentcore](), [amazon-web-services-aws](), [automated-reasoning](), [aws](), [developer](), [formal-verification](), [generative-ai](), [large-language-models-llms](), [llm](), [programming](), [programming-language](), [research](), [responsible-ai](), [safety](), [software-testing](), [testing](), [trustworthy-ai]() ### AI overview Amazon is making a long-term investment in the Lean Focused Research Organization to advance Lean, a programming language for mathematical correctness proofs. The article highlights Lean-based verification for safer AI agents and AWS systems. ### Source excerpt As AI agents take on higher-stakes decisions, Lean programming language makes it possible to mathematically prove they will behave safely. ## Amazon and University of Michigan give robots a sense of touch DevFeed: [Amazon and University of Michigan give robots a sense of touch]() Original publisher: [Read original article]() Author: Mani Nambi; Nima Fazeli Published: 2026-07-10T17:13:31Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [contact-rich manipulation]() Tags: [amazon](), [amazon-robotics-research](), [contact-rich-manipulation](), [dataset-development](), [dexterous-robot-manipulation](), [gelsight-mini-sensor](), [hydroelastic-contact-model](), [hydroshear-simulator](), [physics](), [reinforcement-learning](), [reinforcement-learning-robotics](), [robot-grasping-simulation](), [robot-sense-of-touch](), [robotic-manipulation](), [robotics](), [robots](), [sim-to-real-transfer-robotics](), [training]() ### AI overview HydroShear is a physics-based tactile-force simulator that trains robots for dexterous, contact-rich manipulation in simulation and transfers the resulting policies to real-world tasks. ### Source excerpt HydroShear, a new physics-based simulator, teaches robots how to use their sense of touch to perform complex manipulation tasks, in a way that transfers seamlessly to the real world. ## Capturing token IDs during agentic interactions for better reinforcement learning DevFeed: [Capturing token IDs during agentic interactions for better reinforcement learning]() Original publisher: [Read original article]() Author: Frederick Robinson Published: 2026-07-09T12:46:00Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Reinforcement learning](), [Agent Harness](), [tokenization](), [Language models](), [Rust](), [computer-use](), [multimodal]() Tags: [agent](), [agentic](), [agentic-ai](), [amazon-agi-lab](), [computer-use](), [dataset-development](), [language-models](), [machine-learning](), [model](), [multimodal](), [reinforcement-learning](), [rust](), [tokens](), [training]() ### AI overview This article explains how Turnstile, a Rust proxy placed between an agent harness and a model backend, captures exact token IDs during generation. The recorded token-level trajectories preserve information that text transcripts can lose and can be passed into reinforcement-learning training stacks. Reported validations cover a text-only coding agent and a multimodal computer-use agent whose performance improved during RL runs. ### Source excerpt A new Rust proxy called Turnstile sits between the model backend and the agent harness to capture information lost in mere text transcripts. ## How Amazon tracks carbon intensity across its operations DevFeed: [How Amazon tracks carbon intensity across its operations]() Original publisher: [Read original article]() Author: Kerry Constabile Published: 2026-07-01T15:56:42Z Content type: news Language: en Sources: [Amazon Science homepage]() Topics: [amazon](), [data](), [Cloud](), [Routing (disambiguation)]() Tags: [amazon](), [amazon-sustainability-report](), [cloud-services](), [data](), [decarbonization](), [energy](), [operations](), [renewable-energy](), [routing](), [scale](), [sustainability](), [the-climate-pledge](), [transportation]() ### AI overview Amazon describes its use of sector-specific carbon-intensity metrics to track decarbonization across diverse operations. For retail, it measures emissions per unit shipped and reports a 39% reduction from 2019 to the end of 2025, attributing progress to carbon-free energy, routing improvements, lighter packaging, lower-carbon fuels, alternative transportation, and electric vehicles. ### Source excerpt Amazon is developing precise, sector-specific approaches to measuring decarbonization progress -- starting with emissions per unit shipped. ## The fuel of the future is already here: Why TRISO matters DevFeed: [The fuel of the future is already here: Why TRISO matters]() Original publisher: [Read original article]() Author: Katy Huff Published: 2026-06-24T19:57:09Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [AI Infrastructure](), [amazon](), [cloud-computing]() Tags: [advanced-nuclear-reactors](), [advanced-reactor-fuel-fabrication](), [ai-infrastructure](), [amazon](), [amazon-nuclear-energy](), [clean-energy-data-centers](), [cloud-computing](), [decarbonization](), [durability](), [energy](), [haleu](), [integrity](), [modular-nuclear-reactors](), [nuclear-energy-for-ai](), [physical-science](), [renewable-energy](), [resilience](), [safety](), [science](), [sustainability](), [triso-fuel](), [triso-particles](), [x-energy-xe-100]() ### AI overview Amazon explains how TRISO nuclear fuel particles use layered carbon and ceramic coatings to contain radioactive byproducts and withstand extreme temperatures. The article connects the technology to rising energy demands from AI infrastructure and cloud computing, citing testing that found no detectable failures at 1600°C for 300 hours. ### Source excerpt Millimeter-scale particles of nuclear-reactor fuel are encased in four layers of different materials that act as a "miniature containment system". ## EC2's formally verified "isolation engine" provides mathematical assurance of virtual-machine isolation DevFeed: [EC2's formally verified "isolation engine" provides mathematical assurance of virtual-machine isolation]() Original publisher: [Read original article]() Author: Dominic Mulligan; Nathan Chong Published: 2026-06-10T15:00:00Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Amazon EC2](), [Rust formal verification]() Tags: [amazon-web-services-aws](), [autocorrode-library](), [automated-reasoning](), [aws-graviton5-security](), [cloud-and-systems](), [confidentiality-integrity-proofs](), [ec2-virtual-machine-security](), [formal-verification](), [formally-verified-hypervisor](), [isabelle-hol-proof-assistant](), [kernel](), [memory-safety-verification](), [nitro-isolation-engine](), [nitro-system](), [provable-security](), [rust](), [rust-formal-verification](), [security](), [security-privacy-and-abuse-prevention]() ### AI overview Amazon describes formally verifying the Nitro Isolation Engine, a separation-kernel component that enforces isolation between EC2 virtual machines. The verification uses Isabelle/HOL and is deployed as an always-on feature for Graviton5 users. ### Source excerpt Splitting the "separation kernel" off from the rest of the Nitro security system and using only a subset of the Rust programming language to code it enabled its formal verification. ## Graviton5's improved design increases speed and energy efficiency -- beyond Moore's law DevFeed: [Graviton5's improved design increases speed and energy efficiency -- beyond Moore's law]() Original publisher: [Read original article]() Author: Ali Saidi Published: 2026-06-10T15:00:00Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [cpu](), [Amazon EC2]() Tags: [amazon-elastic-compute](), [architecture](), [cache](), [chip-design](), [cloud-and-systems](), [cpu](), [design](), [energy-efficiency](), [formal-verification](), [graviton](), [hardware](), [memory](), [performance](), [speed]() ### AI overview Amazon describes Graviton5 CPU and M9g/M9gd EC2 instances, highlighting more cores, faster memory and interconnects, improved branch prediction, and expanded cache capacity. ### Source excerpt A new chiplet architecture, custom die-to-die connectivity, and support for DDR5-8800 memory and the latest PCIe gen6 interconnects improve performance by 25% for general-purpose and agentic AI workloads. ## Real-world grounding in agentic AI DevFeed: [Real-world grounding in agentic AI]() Original publisher: [Read original article]() Author: Rose Yu Published: 2026-06-08T19:00:00Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [AI Bots](), [dashboards]() Tags: [agentic-ai](), [agentic-ai-grounding](), [ai](), [ai-agent-hallucinations](), [ai-agents](), [calibrated-uncertainty](), [formal-verification-ai](), [foundation-models-physical-ai](), [hallucinations](), [large-language-models-llms](), [machine-learning](), [physical-ai](), [physics-guided-deep-learning](), [robotics](), [robots](), [uncertainty-quantification-llms]() ### AI overview The article proposes four approaches for grounding AI agents in physical-world operational environments. It argues that integrating domain data, physical principles, and simulations can reduce harmful hallucinations and improve safe, trustworthy agent behavior. ### Source excerpt Four approaches can dramatically improve the performance and trustworthiness of AI agents in operational environments. ## Bridging intent and execution in agentic systems DevFeed: [Bridging intent and execution in agentic systems]() Original publisher: [Read original article]() Author: Gaurav Gupta; Vatshank Chaturvedi Published: 2026-06-08T17:00:00Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Agent Harness](), [AI Agent](), [systems](), [Large Language Model](), [Terminal](), [cloud-infrastructure](), [Open Source](), [Code]() Tags: [agent](), [agentic](), [agentic-ai](), [ai](), [ai-agent](), [benchmarks](), [cloud-and-systems](), [code](), [conversational-ai](), [infrastructure](), [llm](), [software](), [tools]() ### AI overview The article argues that performance in agentic systems is fundamentally a systems problem involving the interaction between a large language model and its harness. It defines the intent-execution gap between model intentions and harness actions, shows that reducing this gap can improve benchmark results without task-specific tuning, and emphasizes the effects of tools, execution graphs, infrastructure, timeouts, and resource constraints. It also introduces Simple Strands Agent (SSA), a lightweight customizable harness, and argues that model-harness codesign is important because model families differ in tool use and feedback interpretation. ### Source excerpt The harnesses that mediate between models and tools in agentic systems are becoming their own performance bottleneck, but a few simple design principles can fix what ails them. ## Ground truth is a process, not a dataset DevFeed: [Ground truth is a process, not a dataset]() Original publisher: [Read original article]() Author: Venkatesh Saligrama Published: 2026-06-03T15:56:57Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [ai observability](), [AI Chat]() Tags: [ai](), [ai-fact-checking](), [ai-generated-research-reports](), [audit-then-score](), [benchmarking](), [conversational-ai](), [datasets](), [deep-research-verification](), [deepfact-bench](), [evaluation](), [fact-verification](), [fact-verification-benchmark](), [generative-ai](), [ground-truth-benchmark-quality](), [hallucination-detection](), [hallucinations](), [human-ai-evaluation](), [large-language-models-llms](), [llm-evaluation-benchmarking](), [long-context](), [machine-learning](), [responsible-ai](), [retrieval-augmented-generation-rag]() ### AI overview The article argues that evaluating factuality in long AI-generated research reports requires a process-based approach to ground truth. It introduces audit-then-score and accompanying datasets for benchmarking AI fact checkers. ### Source excerpt Automatically fact-checking long, AI-generated research reports poses new challenges -- including benchmarking. ## How flat is replacing fat in AWS data center networks DevFeed: [How flat is replacing fat in AWS data center networks]() Original publisher: [Read original article]() Author: Giacomo Bernardi; Ratul Mahajan; Seshadhri Comandur Published: 2026-05-28T10:30:00Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [networking](), [Network architectures]() Tags: [aws](), [aws-data-center-network-architecture](), [cloud-and-systems](), [data-center](), [fat-tree-network-replacement](), [flat-network-topology](), [network-design](), [networking](), [networks](), [quasi-random-network-topology](), [resilience](), [routing]() ### AI overview The article describes AWS's scalable flat data-center network, using quasi-random topology and ShuffleBoxes to replace conventional fat-tree designs. ### Source excerpt "Quasi-random" network topologies and new passive optical components called ShuffleBoxes make more-efficient flat networks as practical as traditional "fat-tree" networks. ## Amazon Research Awards recipients announced DevFeed: [Amazon Research Awards recipients announced]() Original publisher: [Read original article]() Author: Amazon Research Awards team Published: 2026-05-27T17:21:51Z Content type: news Language: en Sources: [Amazon Science homepage]() Topics: [Security, Privacy and Abuse Prevention](), [Cryptography]() Tags: [agentic-ai](), [ai-ml](), [amazon](), [amazon-research-awards](), [automated-reasoning](), [aws](), [cryptography](), [cybersecurity](), [datasets](), [internal-ara-program-updates](), [research]() ### AI overview Amazon announced 70 Fall 2025 Research Awards recipients from 49 universities in 11 countries. The funded calls span AI for information security, agentic AI, automated reasoning, cryptography, cybersecurity and anti-abuse technologies, and sustainability. ### Source excerpt Awardees represent more than 49 universities in 11 countries. Recipients have access to Amazon public datasets, along with AWS AI/ML services and tools. ## Diverse reasoning traces teach LLMs to make better decisions DevFeed: [Diverse reasoning traces teach LLMs to make better decisions]() Original publisher: [Read original article]() Author: Sheng Jia; Xiao Wang; Shiva Kasiviswanathan Published: 2026-05-26T15:17:06Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Language models]() Tags: [benchmarks](), [chain-of-thought](), [code-generation](), [conversational-ai](), [deep-learning](), [evaluation](), [fine-tuning](), [iclr](), [iclr-2026](), [large-language-models-llms](), [llms](), [math-reasoning](), [parallel-reasoning](), [post-training](), [post-training-optimization](), [reasoning](), [reinforcement-learning](), [traces]() ### AI overview The article presents set-supervised fine tuning and global forking policy optimization to train LLMs on multiple distinct reasoning paths. It reports 5% to 7% single-shot accuracy gains on standard benchmarks. ### Source excerpt How to train language models to generate diverse, accurate reasoning paths using tokens that control distinct reasoning strategies. ## Making LLMs faster without sacrificing accuracy DevFeed: [Making LLMs faster without sacrificing accuracy]() Original publisher: [Read original article]() Author: Tao Yu; Youngsuk Park Published: 2026-05-15T13:00:00Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Language models](), [Training AI Models](), [web applications]() Tags: [architecture](), [chinchilla-scaling-law](), [conversational-ai](), [generative-ai](), [grouped-query-attention](), [hyperparameter-optimization](), [iclr-2026](), [inference](), [inference-efficiency](), [large-language-models](), [large-language-models-llms](), [llm-optimization](), [llms](), [machine-learning](), [model-architecture](), [network-architectures](), [scaling-laws](), [training](), [transformer-architecture]() ### AI overview The article presents scaling laws that connect LLM architectural choices to the tradeoff between accuracy and efficiency. It describes how these choices can improve inference throughput without reducing accuracy. ### Source excerpt A new scaling law that relates particular architectural choices to loss helps identify models that improve throughput by up to 47% with no loss of accuracy. ## Promptimus: Improving already good LLM prompts with zero manual engineering DevFeed: [Promptimus: Improving already good LLM prompts with zero manual engineering]() Original publisher: [Read original article]() Author: Zhengyuan Shen; Yunfei Bai; Sullam Jeoung; Shuai Wang Published: 2026-05-14T13:47:45Z Content type: article Language: en Sources: [Amazon Science homepage]() Topics: [Automated prompt engineering](), [Prompt Engineering](), [Optimization](), [Large Language Model](), [AI Agent](), [Artificial Intelligence](), [multimodal](), [Code]() Tags: [ai](), [ai-agent](), [amazon-bedrock](), [automated-prompt-engineering](), [code-generation](), [conversational-ai](), [engineering](), [generative-ai](), [large-language-models-llms](), [llm](), [multimodal](), [optimization](), [performance](), [prompt-engineering](), [prompt-optimization]() ### AI overview Promptimus is an automated prompt-engineering method for improving already strong prompts without manual engineering. It uses task data, user-defined performance metrics, failure analysis, debugging agents, sanitization, and targeted edit mode to refine prompts while preserving working business logic. The method is model agnostic and supports textual and multimodal LLM tasks, including classification, extraction, generation, summarization, code generation, and tool use. ### Source excerpt By focusing on specific failure points and suggesting targeted solutions, a new automated prompt-engineering framework improves prompt performance without compromising existing functionality. [Next page]()