# Sequences

A sequence is a function that maps indexed natural numbers to elements of a set, and may be finite or infinite; sequences are also called arrays, vectors, or lists.

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

## Exact Sequence Interpolation with Transformers

DevFeed: [Exact Sequence Interpolation with Transformers](<https://devfeed.tech/articles/exact-sequence-interpolation-with-transformers-141517.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2502.02270>)

Author: Albert Alcalde, Giovanni Fantuzzi, Enrique Zuazua

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

Content type: article

Language: en

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

Topics: [Transformers](<https://devfeed.tech/topics/transformers.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [attention](<https://devfeed.tech/tags/attention.md>), [blocks](<https://devfeed.tech/tags/blocks.md>), [clustering](<https://devfeed.tech/tags/clustering.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [convergence](<https://devfeed.tech/tags/convergence.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [global](<https://devfeed.tech/tags/global.md>), [learning](<https://devfeed.tech/tags/learning.md>), [low-rank](<https://devfeed.tech/tags/low-rank.md>), [math-oc](<https://devfeed.tech/tags/math-oc.md>), [models](<https://devfeed.tech/tags/models.md>), [performance](<https://devfeed.tech/tags/performance.md>), [self-attention](<https://devfeed.tech/tags/self-attention.md>), [sequence](<https://devfeed.tech/tags/sequence.md>), [softmax](<https://devfeed.tech/tags/softmax.md>), [stat-ml](<https://devfeed.tech/tags/stat-ml.md>), [theoretical](<https://devfeed.tech/tags/theoretical.md>), [training](<https://devfeed.tech/tags/training.md>), [transformers](<https://devfeed.tech/tags/transformers.md>)

### AI overview

The paper constructs transformers that exactly interpolate datasets of finite input sequences in ℝ^d (d >= 2) when each output sequence is no longer than its input. For N examples with output lengths m^j, the construction uses O(Σ_j m^j) blocks and O(d Σ_j m^j) parameters, with bounds independent of input sequence length. It alternates feed-forward and self-attention layers, uses self-attention's clustering effect, and employs low-rank attention matrices. The authors first establish the result for hardmax attention, then extend it to softmax. They also report convergence guarantees to a global minimizer for learning problems under regularized training strategies.

### Source excerpt

arXiv:2502.02270v4 Announce Type: replace Abstract: We prove that transformers can exactly interpolate datasets of finite input sequences in $\mathbb{R}^d$, $d\geq 2$, with corresponding output sequences of smaller or equal length. Specifically, given $N$ sequences of arbitrary but finite lengths in $\mathbb{R}^d$ and output sequences of lengths $m^1, \dots, m^N \in \mathbb{N}$, we construct a transformer with $\mathcal{O}(\sum_{j=1}^N m^j)$ blocks and $\smash{\mathcal{O}(d \sum_{j=1}^N m^j)}$ parameters that exactly interpolates the dataset. Our construction provides complexity estimates that are independent of the input sequence length, by alternating feed-forward and self-attention layers and by capitalizing on the clustering effect inherent to the latter. Our novel constructive method also uses low-rank parameter matrices in the self-attention mechanism, a common feature of practical transformer implementations. These results are first established in the hardmax self-attention setting, where the geometric structure permits an explicit and quantitative analysis, and are then extended to the softmax setting. Finally, we demonstrate the applicability of our exact interpolation construction to learning problems, in particular by providing convergence guarantees to a global minimizer under regularized training strategies. Our analysis contributes to the theoretical understanding of transformer models, offering an explanation for their excellent performance in exact sequence-to-sequence interpolation tasks.

## Halting in Recurrent Sequence Models

DevFeed: [Halting in Recurrent Sequence Models](<https://devfeed.tech/articles/halting-in-recurrent-sequence-models-139037.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20377>)

Author: Simon Dr\\"ager

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

Content type: article

Language: en

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

Topics: [AI Models](<https://devfeed.tech/topics/ai-models.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [alignment](<https://devfeed.tech/tags/alignment.md>), [boundary](<https://devfeed.tech/tags/boundary.md>), [copy](<https://devfeed.tech/tags/copy.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [encoding](<https://devfeed.tech/tags/encoding.md>), [eos](<https://devfeed.tech/tags/eos.md>), [halting-problem](<https://devfeed.tech/tags/halting-problem.md>), [hidden-states](<https://devfeed.tech/tags/hidden-states.md>), [linear](<https://devfeed.tech/tags/linear.md>), [neural-networks](<https://devfeed.tech/tags/neural-networks.md>), [number](<https://devfeed.tech/tags/number.md>), [phases](<https://devfeed.tech/tags/phases.md>), [recurrent](<https://devfeed.tech/tags/recurrent.md>), [repeat](<https://devfeed.tech/tags/repeat.md>), [rotation](<https://devfeed.tech/tags/rotation.md>), [sequence](<https://devfeed.tech/tags/sequence.md>), [state-space-models](<https://devfeed.tech/tags/state-space-models.md>), [transformation](<https://devfeed.tech/tags/transformation.md>)

### AI overview

In a task where models copy an input a specified number of times before emitting EOS, tanh RNNs and linear SSMs show different candidate mechanisms for deciding when to stop. RNN hidden states encode the remaining tokens in a low-dimensional linear subspace; states with the same remaining count converge despite different instructed repeat counts. An affine state shift at a sequence boundary moved the RNN to the representation for one fewer copy, causing it to stop early while producing the shortened target on 94.4% of new sequences. In linear SSMs, repeat-count differences occupy two-dimensional rotation planes whose contributions to EOS comparison cancel during copying and align near stopping. Independently rotating the planes disrupted most intended stopping while preserving copied content, implicating their relative phases. These results concern a minimal copying task and identify distinct candidate halting primitives in the two model types.

### Source excerpt

arXiv:2611.20377v1 Announce Type: new Abstract: Recurrent sequence models, such as recurrent neural networks (RNNs) and state space models (SSMs), are often trained to emit an end-of-sequence (EOS) token when their output sequence is complete. During training, EOS is treated as another token without a special loss term or incentive. Although this training paradigm is widely used, it is not understood how models decide to output EOS. We study a minimal version of the EOS halting problem in which networks are trained to emit a given input sequence an instructed number of times, followed by EOS. We train tanh RNNs and linear SSMs on this task, finding that the RNNs encode the number of tokens left to copy in a linear subspace, of which two principal components capture most of the variation. Hidden states with the same remaining count converge to a shared representation regardless of their instructed repeat count. By modulating the hidden state using an affine transformation at the boundary between copied sequences, we shift the RNN's state to the representation shared by states with one fewer copy remaining. The unchanged network emits EOS one copy earlier while reproducing the shortened target sequence in 94.4% of new sequences, revealing that this representation causally determines the stopping time. In the SSMs, the difference between trajectories with different repeat counts is separated across two-dimensional rotation planes whose contributions to the EOS comparison cancel during copying and align near the stopping time. Rotating the planes independently removes most of the intended stopping while preserving the copied content, which shows that the stopping time depends on the planes' relative phases. These results identify two candidate primitives of EOS halting in recurrent models: a shared encoding of the number of remaining tokens in tanh RNNs and phase alignment across fixed rotations in linear SSMs.

## EvoFlows: Evolutionary Edit-Based Flow-Matching for Protein Engineering

DevFeed: [EvoFlows: Evolutionary Edit-Based Flow-Matching for Protein Engineering](<https://devfeed.tech/articles/evoflows-evolutionary-edit-based-flow-matching-for-protein-engineering-132767.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.11703>)

Author: Nicolas Deutschmann, Constance Ferragu, Jonathan D. Ziegler, Shayan Aziznejad, Eli Bixby

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

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

Topics: [Sequences](<https://devfeed.tech/topics/sequences.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [binder](<https://devfeed.tech/tags/binder.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [flow-matching](<https://devfeed.tech/tags/flow-matching.md>), [functional](<https://devfeed.tech/tags/functional.md>), [mutation](<https://devfeed.tech/tags/mutation.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [proteins](<https://devfeed.tech/tags/proteins.md>), [sequence](<https://devfeed.tech/tags/sequence.md>), [stable](<https://devfeed.tech/tags/stable.md>), [validation](<https://devfeed.tech/tags/validation.md>), [variable](<https://devfeed.tech/tags/variable.md>)

### AI overview

EvoFlows models protein optimization as variable-length edit flows between evolutionarily related proteins, predicting which mutations to make, where to make them, and how many to introduce. This supports insertions and deletions without pre-specified mutation sites, unlike the autoregressive, masked-language, and discrete-diffusion approaches described in the abstract. Across UniRef and OAS protein families, the authors report variants that remain family-consistent while exploring farther from the template than leading baselines. In vitro tests of a VHH binder and an enzyme found that most generated variants remained functional and stable without computational downselection.

### Source excerpt

arXiv:2603.11703v3 Announce Type: replace Abstract: Protein optimization improves an existing protein through targeted mutations, but existing protein language models are poorly suited to it: autoregressive models require full sequence generation, masked language and discrete diffusion models rely on pre-specified mutation locations, and none naturally support insertions and deletions relative to a template. We introduce EvoFlows, a variable-length sequence-to-sequence approach that learns mutational trajectories between evolutionarily related proteins via edit flows, predicting not only _which_ mutation to perform, but also _where_ and _how many_ should occur. Across diverse protein families from UniRef and OAS, EvoFlows generates variants that remain consistent with natural families while exploring farther from the template than leading baselines. _In vitro_ validation of a VHH binder and an enzyme confirms that the large majority of generated variants remain functional and stable without computational downselection.

## TGCM: Topic-Guided Consistency Modeling for One-Step Disentanglement of Interleaved APT Technique Sequences

DevFeed: [TGCM: Topic-Guided Consistency Modeling for One-Step Disentanglement of Interleaved APT Technique Sequences](<https://devfeed.tech/articles/tgcm-topic-guided-consistency-modeling-for-one-step-disentanglement-of-interleaved-apt-technique-sequences-128886.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2606.18651>)

Author: Guo-Wei Wong, Ming-Chuan Yang, Shou-De Lin, Wang-Chien Lee, Meng Chang Chen

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

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Threat monitoring](<https://devfeed.tech/topics/threat-monitoring.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [abstraction](<https://devfeed.tech/tags/abstraction.md>), [advanced](<https://devfeed.tech/tags/advanced.md>), [advanced-persistent-threat](<https://devfeed.tech/tags/advanced-persistent-threat.md>), [also](<https://devfeed.tech/tags/also.md>), [apt](<https://devfeed.tech/tags/apt.md>), [artifacts](<https://devfeed.tech/tags/artifacts.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [attack](<https://devfeed.tech/tags/attack.md>), [audit](<https://devfeed.tech/tags/audit.md>), [audit-logs](<https://devfeed.tech/tags/audit-logs.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [campaign](<https://devfeed.tech/tags/campaign.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [modeling](<https://devfeed.tech/tags/modeling.md>), [one](<https://devfeed.tech/tags/one.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [topic](<https://devfeed.tech/tags/topic.md>)

### AI overview

TGCM addresses the problem of separating concurrent APT campaigns from interleaved audit events when the number of campaigns is unknown. It treats demixing as assigning individual MITRE ATT&CK technique occurrences and reconstructing campaign sequences, doing both in one forward pass. The method combines guidance from ATT&CK narratives with an embedding-space self-consistency objective. Across synthetic and mixed benchmarks, DARPA traces, and the CAPTure benchmark, the abstract reports improved occurrence assignment under heavy interleaving, repeated technique reuse, symbolic extraction errors, and budgeted unknown-campaign inference. It also reports generalization to unseen benchmarks without retraining and effectiveness with practical ATT&CK extraction pipelines.

### Source excerpt

arXiv:2606.18651v3 Announce Type: replace Abstract: Multiple Advanced Persistent Threat (APT) campaigns may execute concurrently, producing audit logs whose events are interleaved without explicit campaign boundaries. We formulate this setting as Unknown-K Interleaved Sequence Demixing (UKISD): recovering coherent campaign episodes when the number of concurrent campaigns is unknown. Each episode is represented as a sequence of MITRE ATT&CK technique occurrences, and demixing is cast as occurrence-level assignment with sequence reconstruction. Existing methods often assume single-campaign observations or rely on local heuristics, limiting robustness to severe interleaving, repeated techniques, abstraction noise, and unknown mixture cardinality. We propose Topic-Guided Consistency Modeling (TGCM), a consistency-inspired one-step framework for UKISD. TGCM maps an interleaved technique sequence to campaign episodes, jointly performing occurrence assignment and sequence reconstruction in one forward pass. It combines topic guidance from ATT&CK narratives with an embedding-space self-consistency objective to improve semantic coherence and retain efficient inference. We evaluate TGCM on synthetic mixtures, mixed benchmarks, DARPA engagement traces, and CAPTure, an end-to-end benchmark containing 200 scenarios, 855 million audit events, and 25 ATT&CK-aligned attack profiles in single-host and multi-host settings. TGCM improves occurrence-level assignment under heavy interleaving, repeated technique reuse, symbolic extraction errors, and budgeted unknown-K inference. It also generalizes to unseen benchmarks without retraining and remains effective with practical ATT&CK extraction pipelines. These results establish TGCM as a practical post-abstraction framework for reasoning about concurrent APT campaigns. Code and artifacts are available at https://irish-kw.github.io/TGCM_Website/.

## INTENT: Out-of-Sample, Out-of-Time, and Cold-Start Evaluation on User Event Sequences

DevFeed: [INTENT: Out-of-Sample, Out-of-Time, and Cold-Start Evaluation on User Event Sequences](<https://devfeed.tech/articles/intent-out-of-sample-out-of-time-and-cold-start-evaluation-on-user-event-sequences-128652.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.13069>)

Author: Mattias Kvist, Erik Lidman Hillbom, Edoardo de Cal, Davis Siemens, Adam Fredriksson, Rickard Br\\"uel-Gabrielsson

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

Content type: article

Language: en

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

Topics: [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Dataset development](<https://devfeed.tech/topics/dataset-development.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cold-start](<https://devfeed.tech/tags/cold-start.md>), [cs-ir](<https://devfeed.tech/tags/cs-ir.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [event](<https://devfeed.tech/tags/event.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [time](<https://devfeed.tech/tags/time.md>), [user](<https://devfeed.tech/tags/user.md>)

### AI overview

INTENT benchmarks next-action prediction across seven cells defined by whether the user, time period, and target item appeared in training. An audit of 59 public dataset families retained five; missing calendar timestamps were the most common defect, and each dataset uses a cutoff chosen by a rule set before model evaluation. In reference results, a transformer performed best on seen items after the cutoff but scored zero on new items. Item metadata helped mainly on MicroLens, while repeating the user's last item led on new items in four datasets. That parameter-free baseline indicates that on three datasets most post-cutoff score loss reflects changed behavior rather than model failure. The benchmark pipeline, cutoff rule, and per-cell user counts are public.

### Source excerpt

arXiv:2611.13069v1 Announce Type: new Abstract: Banks now train foundation models on ordered sequences of customer events such as transactions, app sessions, and authorizations. Deployed, such a model predicts forward in time, often for customers and merchants absent from training, yet each is evaluated on private data, so no two can be compared. Public benchmarks for such sequences rarely test a period the model has not seen; recommendation studies usually hold out each user's last events instead. We present INTENT (INTEraction, Novelty, and Transfer evaluation for next-action prediction), a public benchmark that scores a model in seven cells defined by whether the user, the period, and the target item appeared in training. An audit of 59 public dataset families retains five datasets; the most common defect is the lack of a calendar timestamp that orders events. No date lies within the time span of all five datasets, so each is split at its own cutoff, chosen by one rule fixed before any model is evaluated. Summed over the datasets, every cell holds at least 111,769 users. In reference results the cell determines the best model: a transformer leads on seen items after the cutoff but scores zero on new items, a variant that embeds item metadata helps mainly on MicroLens, and repeating the user's last item leads on new items in four of five datasets. This parameter-free control also shows that on three datasets most of the score lost after the cutoff reflects changed behavior rather than the model. The pipeline, the cutoff rule, and the per-cell user counts are publicly available.

## How Finite Brains Generalize Without Bound: Spatial Memory as a Universal Computer

DevFeed: [How Finite Brains Generalize Without Bound: Spatial Memory as a Universal Computer](<https://devfeed.tech/articles/how-finite-brains-generalize-without-bound-spatial-memory-as-a-universal-computer-120268.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.07920>)

Author: Yuhang He, Si Wu

Published: 2026-11-06T05:00:00Z

Content type: article

Language: en

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

Topics: [compression and generalization](<https://devfeed.tech/topics/compression-and-generalization.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Transformers](<https://devfeed.tech/topics/transformers.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [arithmetic](<https://devfeed.tech/tags/arithmetic.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [brain](<https://devfeed.tech/tags/brain.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cycles](<https://devfeed.tech/tags/cycles.md>), [grammar](<https://devfeed.tech/tags/grammar.md>), [grid](<https://devfeed.tech/tags/grid.md>), [lstm](<https://devfeed.tech/tags/lstm.md>), [machine](<https://devfeed.tech/tags/machine.md>), [memory](<https://devfeed.tech/tags/memory.md>), [network](<https://devfeed.tech/tags/network.md>), [numbers](<https://devfeed.tech/tags/numbers.md>), [position](<https://devfeed.tech/tags/position.md>), [spatial](<https://devfeed.tech/tags/spatial.md>), [transformer](<https://devfeed.tech/tags/transformer.md>), [transformers](<https://devfeed.tech/tags/transformers.md>), [turing-machines](<https://devfeed.tech/tags/turing-machines.md>)

### AI overview

The paper argues that finite-experience learners can generalize to longer inputs when their decisions depend on equality or coincidence relations rather than coordinates such as positions, counts, or times, whose values keep expanding. It proposes memory built from repeating cycles, with place cells detecting cycle returns and a hippocampal index recording past events; a prefrontal controller moves the cycles without reading them directly. The authors connect this design to Turing-complete computation and describe a neural-network implementation. Trained on strings up to 40 symbols, it reportedly reaches at least 99% average accuracy over lengths 41-500 on nine of 15 benchmark tasks, including five unsolved by the tested LSTMs or Transformers. With new cycles added after training, it reverses strings of 15,015 symbols without error, then fails along a curve the theory predicts. The excerpt does not provide further experimental or theoretical detail.

### Source excerpt

arXiv:2611.07920v1 Announce Type: new Abstract: A child who learns to add small numbers adds numbers of any length. LSTMs and Transformers trained on short sequences fail on longer ones. We trace the difference to what a learner reads when it decides. A position, a count or a time keeps taking new values as the input grows: it is a coordinate, and a learner that reads one meets in every long input values its experience never covered. Whether two things are the same has the same two answers at every size: it is a coincidence, and a learner that reads only coincidences finds in a long input nothing that short inputs lacked, so finite experience fixes its rule for every size. The memory carries the size of the input and the rule stays finite, and this determines the memory a generalizing brain can have: a memory that grows while what the learner reads of it stays finite has to look the same from every place, so it is made of cycles, and the past is reached by asking what happened when the cycles last stood where they stand now. The cycles are the grid modules, the cells that detect their return the place cells, the records the hippocampal index; a prefrontal controller moves the cycles and never sees them. Reading only coincidences costs no power: what finite experience fixes in such a learner is what Turing machines compute in step with their input, which is why a circuit that evolved to find the way home can learn arithmetic and grammar. In a network of neurons this machine of cycles is the dynamics seen through its symmetry, and symbols are its invariants. Trained on strings up to 40 symbols, the machine is at least 99% correct, on average over lengths 41 to 500, on nine of fifteen benchmark tasks, five of them tasks that no LSTM or Transformer solves; given new cycles after training it reverses strings of 15,015 symbols without error and fails beyond that on the curve the theory predicts; and its controllers run unchanged on neurons.

## How Far Can Chord-Symbol Time-Series Adaptation Carry Genre Identity? Capabilities and Boundaries in Multi-Genre Chord-Symbol Modeling

DevFeed: [How Far Can Chord-Symbol Time-Series Adaptation Carry Genre Identity? Capabilities and Boundaries in Multi-Genre Chord-Symbol Modeling](<https://devfeed.tech/articles/how-far-can-chord-symbol-time-series-adaptation-carry-genre-identity-capabilities-and-boundaries-in-multi-genre-chord-symbol-modeling-120442.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2606.07334>)

Author: Jinju Lee

Published: 2026-11-06T05:00:00Z

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>), [lora](<https://devfeed.tech/topics/lora.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [adapter](<https://devfeed.tech/tags/adapter.md>), [adapters](<https://devfeed.tech/tags/adapters.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-sd](<https://devfeed.tech/tags/cs-sd.md>), [identity](<https://devfeed.tech/tags/identity.md>), [modeling](<https://devfeed.tech/tags/modeling.md>), [music](<https://devfeed.tech/tags/music.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

Across 11 genres, LoRA, IA3, BitFit, prefix tuning, and full fine-tuning each improve held-out chord prediction over a frozen pop-trained Music Transformer. Much of the gain comes from sequence headers: when those tokens are excluded, gains shrink, and outside Bach chorales four methods perform nearly alike. For most genres, giving the frozen model the pop marker brings it within 0.12 percentage points of the matched adapter; only Bach chorales show a larger adaptation benefit (+9.6 points for LoRA under that comparison). The results suggest chord-symbol adaptation mainly provides a usable genre marker for genres the base already covers, while chord symbols alone do not encode complete genre identity. Perceived genre authenticity remains untested by listener evaluation.

### Source excerpt

arXiv:2606.07334v5 Announce Type: replace-cross Abstract: This report treats chord-symbol sequences as an interpretable, controllable time series for genre-local harmonic modeling. The frozen Music Transformer base is weight-identical to a pop-dominated Phase-0 baseline. I adapt it to eleven target genres: blues, bossa nova, Bach chorales, country, electronic, folk, funk, gospel, hip-hop, R&B/soul, and rock. The main grid compares LoRA, IA3, BitFit, prefix tuning, and full fine-tuning over 11 genres and 3 seeds (165 cells). All five methods improve over the frozen base on held-out chord prediction (macro gains +2.89 to +3.61 percentage points). Top-1 also scores the three header tokens of each sequence, which move whole runs: only a matched adapter predicts its genre marker, and whether a run predicts the placeholder C-major key of every Chordonomicon song is close to arbitrary. Without the header tokens the macro gains are +1.64 to +2.38 pp, and outside Bach chorales LoRA, IA3, prefix tuning, and full fine-tuning give the same gain (+1.30 to +1.34 pp). That gain reflects a handicap of the frozen base, which for every target but rock reads a genre marker it never read in training: given the pop marker instead, it comes within 0.12 pp of the released matched adapter in every non-chorale genre. Wrong-genre adapters beat the frozen base only where it reads such a marker, and outside Bach chorales the matched adapter's edge over the best wrong one is within 0.2 pp. Only on Bach chorales does adaptation add clearly more (+9.6 pp for LoRA over the base given the pop marker). Further diagnostics (rank sweeps, matched-size and same-weights controls, chord-only classification, real songs) support a bounded conclusion: for genres the base already covers, chord-symbol adaptation mainly supplies a usable genre marker, and chord symbols alone do not carry complete genre identity. Perceived genre authenticity is left to controlled listener evaluation.

## Get a load of Excel's new multi-value cells

DevFeed: [Get a load of Excel's new multi-value cells](<https://devfeed.tech/articles/get-a-load-of-excel-s-new-multi-value-cells-116360.md>)

Original publisher: [Read original article](<https://www.computerworld.com/article/4244163/get-a-load-of-excels-new-multi-value-cells.html>)

Author: Howard Wen

Published: 2026-11-04T11:00:00Z

Content type: article

Language: en

Sources: [Computerworld](<https://devfeed.tech/sources/computerworld.md>)

Topics: [Tool](<https://devfeed.tech/topics/tool.md>), [data type](<https://devfeed.tech/topics/data-type.md>), [structure](<https://devfeed.tech/topics/structure.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>), [microsoft 365](<https://devfeed.tech/topics/microsoft-365.md>)

Tags: [array](<https://devfeed.tech/tags/array.md>), [beta-channel](<https://devfeed.tech/tags/beta-channel.md>), [cells](<https://devfeed.tech/tags/cells.md>), [data-structures](<https://devfeed.tech/tags/data-structures.md>), [edition-us](<https://devfeed.tech/tags/edition-us.md>), [excel](<https://devfeed.tech/tags/excel.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [list](<https://devfeed.tech/tags/list.md>), [microsoft-365](<https://devfeed.tech/tags/microsoft-365.md>), [microsoft-365-microsoft-excel-microsoft-office-office-suites-productivity-software](<https://devfeed.tech/tags/microsoft-365-microsoft-excel-microsoft-office-office-suites-productivity-software.md>), [microsoft-office](<https://devfeed.tech/tags/microsoft-office.md>), [nested](<https://devfeed.tech/tags/nested.md>), [office-suites](<https://devfeed.tech/tags/office-suites.md>), [preview](<https://devfeed.tech/tags/preview.md>), [productivity](<https://devfeed.tech/tags/productivity.md>), [productivity-software](<https://devfeed.tech/tags/productivity-software.md>), [scalar](<https://devfeed.tech/tags/scalar.md>), [storytype-tips](<https://devfeed.tech/tags/storytype-tips.md>)

### AI overview

Excel's preview feature lets a single cell hold structured lists, arrays, and nested arrays. Lists preserve separate text, number, date, and Boolean items; arrays hold row-and-column grids, and nested arrays can keep formula results such as TEXTSPLIT outputs inside a cell instead of spilling across the sheet and triggering #SPILL! errors. Create lists with Ctrl+J or Insert > List; braces can contain arrays. Direct references may spill the contents, while INDEX or field access can retrieve selected items. The feature is available on Windows and Mac to Microsoft 365 Insiders in the Beta Channel, but not Excel for the web or perpetual licenses. Limitations include reduced visibility during auditing, higher memory use and potentially slower calculations, compatibility errors in unsupported Excel versions, and limited support from legacy functions such as SUMIF, COUNTIF, and VLOOKUP. Compatibility Version 3 is required for most nested array calculations; preview behavior may change.

### Source excerpt

Until now, Microsoft Excel has followed a strict structural rule: each cell on a worksheet can hold only a single scalar value. While you can string together text using commas or write dynamic array formulas to spill values across adjacent rows and columns, the cell itself remains a single-item container. But Excel has evolved. A feature currently in preview lets you store multiple distinct, structured values within a single cell. Three new data structures can now be put inside a cell: A list maintains a sequence of independent values arranged in a single column. Unlike a plain text string separated by commas, an in-cell list treats each entry as a distinct data element. A list cell can hold a mix of text, numbers, dates, and Boolean values while preserving the native data type of each item. An array stores a grid of rows and columns inside one cell. A nested array allows formulas to generate and return arrays inside other arrays. This allows complex outputs to be confined within individual cells instead of trying to spill them across adjacent worksheet rows and columns. These in-cell data structures can simplify complex formula processing and reduce clutter on the worksheet. Availability: These in-cell data structures are available in Excel for Windows and Mac to members of the Microsoft 365 Insider program who are on the Beta Channel. Microsoft has not yet said when the features will begin rolling out to all M365 subscribers, and they are not supported in perpetual license versions (such as Office 2021 or Office 2024) or Excel for the web. Use cases Here are examples of how these three in-cell structures can be used. Multi-item categorizing Imagine categorizing your co-workers, who each have expertise in using up to five M365 software tools: Excel, OneNote, Outlook, PowerPoint, and Word. Or they specialize in selling any number of categories of products. To store this information on a spreadsheet, you could designate a column for each software tool or product cate

## Using the Python zip() Function for Parallel Iteration

DevFeed: [Using the Python zip() Function for Parallel Iteration](<https://devfeed.tech/articles/using-the-python-zip-function-for-parallel-iteration-110733.md>)

Original publisher: [Read original article](<https://realpython.com/python-zip-function/>)

Author: Leodanis Pozo Ramos

Published: 2026-10-31T14:00:00Z

Content type: tutorial

Language: en

Sources: [Real Python](<https://devfeed.tech/sources/real-python.md>)

Topics: [Python](<https://devfeed.tech/topics/python.md>), [iteration](<https://devfeed.tech/topics/iteration.md>), [function](<https://devfeed.tech/topics/function.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>), [Parameter](<https://devfeed.tech/topics/parameter.md>)

Tags: [build](<https://devfeed.tech/tags/build.md>), [dict](<https://devfeed.tech/tags/dict.md>), [iteration](<https://devfeed.tech/tags/iteration.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [python](<https://devfeed.tech/tags/python.md>), [zip](<https://devfeed.tech/tags/zip.md>)

### AI overview

Python's `zip()` returns a lazy iterator of tuples pairing items from multiple iterables in order, making it useful for parallel loops and calculations. By default, iteration stops at the shortest input, so unmatched items are ignored; `itertools.zip_longest()` instead continues to the longest input and fills gaps with a chosen value. Since Python 3.10, `zip(..., strict=True)` raises `ValueError` when lengths differ. The tutorial also shows how `zip(*pairs)` separates paired sequences and how `dict(zip(keys, values))` creates a dictionary. Ordered sequences and dictionaries preserve meaningful pairings, while sets do not guarantee order.

### Source excerpt

Learn how to use the Python zip() function to iterate over multiple iterables in parallel, unzip sequences, and build dictionaries with less code.

## Python range(): Represent Numerical Ranges

DevFeed: [Python range(): Represent Numerical Ranges](<https://devfeed.tech/articles/python-range-represent-numerical-ranges-109905.md>)

Original publisher: [Read original article](<https://realpython.com/python-range/>)

Author: Geir Arne Hjelle

Published: 2026-10-30T14:00:00Z

Content type: tutorial

Language: en

Sources: [Real Python](<https://devfeed.tech/sources/real-python.md>)

Topics: [Python](<https://devfeed.tech/topics/python.md>), [iteration](<https://devfeed.tech/topics/iteration.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>), [NumPy](<https://devfeed.tech/topics/numpy.md>), [Learning](<https://devfeed.tech/topics/learning.md>)

Tags: [arguments](<https://devfeed.tech/tags/arguments.md>), [built-in](<https://devfeed.tech/tags/built-in.md>), [class](<https://devfeed.tech/tags/class.md>), [constructor](<https://devfeed.tech/tags/constructor.md>), [enumerate](<https://devfeed.tech/tags/enumerate.md>), [function](<https://devfeed.tech/tags/function.md>), [loops](<https://devfeed.tech/tags/loops.md>), [python](<https://devfeed.tech/tags/python.md>), [range](<https://devfeed.tech/tags/range.md>), [zip](<https://devfeed.tech/tags/zip.md>)

### AI overview

Python's `range` creates a lazy, reusable sequence of integers from an inclusive start to an exclusive stop, with an optional nonzero step. It accepts one, two, or three integer-like arguments; a negative step counts downward, and `list()` can materialize the values. Use it for fixed-count repetition or when iterating over numerical values. For ordinary sequence loops, iterate over elements directly; use `enumerate()` when you need indices and elements, and `zip()` to traverse multiple iterables together. Ranges support indexing, slicing, membership checks, and `len()`. `reversed()` provides reverse iteration but returns an iterator without range indexing or length; a reversed slice such as `[::-1]` retains a range and supports varied step sizes. Floating-point ranges require another approach, such as a custom class or NumPy's `arange()`.

### Source excerpt

Master the Python range() function, see how it works under the hood, and learn when enumerate() or zip() beat ranges in your loops.

## DES-JEPA: Learning a Shared Vocabulary for High-Dimensional Discrete Event Sequences

DevFeed: [DES-JEPA: Learning a Shared Vocabulary for High-Dimensional Discrete Event Sequences](<https://devfeed.tech/articles/des-jepa-learning-a-shared-vocabulary-for-high-dimensional-discrete-event-sequences-108080.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.38445>)

Author: Hugo Math, Maximilian Forstenh\\"ausler, Christos Anagnostopoulos

Published: 2026-10-30T04:00:00Z

Content type: article

Language: en

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

Topics: [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [challenge](<https://devfeed.tech/tags/challenge.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [context](<https://devfeed.tech/tags/context.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [des](<https://devfeed.tech/tags/des.md>), [domain](<https://devfeed.tech/tags/domain.md>), [domains](<https://devfeed.tech/tags/domains.md>), [embedding](<https://devfeed.tech/tags/embedding.md>), [event](<https://devfeed.tech/tags/event.md>), [event-streams](<https://devfeed.tech/tags/event-streams.md>), [events](<https://devfeed.tech/tags/events.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [foundation](<https://devfeed.tech/tags/foundation.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [framework](<https://devfeed.tech/tags/framework.md>), [inference](<https://devfeed.tech/tags/inference.md>), [learning](<https://devfeed.tech/tags/learning.md>), [sequences](<https://devfeed.tech/tags/sequences.md>)

### AI overview

DES-JEPA and DES-PFN address discrete event sequences whose domain-specific labels do not align and whose large vocabularies exceed limits of current DES foundation models. DES-JEPA learns a shared discrete vocabulary from synthetic stochastic processes with unaligned vocabularies; DES-PFN reuses it for in-context next-event prediction. The authors report competitive zero-shot performance without task-specific fine-tuning on eight unseen real-world datasets, scaling to 700 event types.

### Source excerpt

arXiv:2610.38445v1 Announce Type: new Abstract: Temporal foundation models such as Chronos and TimesFM generalize across domains through a shared numerical observation space. In contrast, discrete event sequences (DESs), including operational event logs or patient clinical-event trajectories, have domain-specific event vocabularies whose labels lack a consistent correspondence across domains. Beyond this lack of consistency, the number of distinct events in real-world settings is high enough to challenge both classical temporal point process modeling and current in-context foundation inference models such as FIM-PP. Critically, current foundation models for DESs are limited to a small, fixed number of event types, restricting their applicability to realistic, high-dimensional event streams. To address this, we propose two complementary modules: (i) DES-JEPA, a joint-embedding predictive architecture that learns a shared discrete vocabulary from synthetic stochastic processes with unaligned event vocabularies; and (ii) DES-PFN, a discrete event sequence prior-fitted network that reuses the learned vocabulary for in-context next-event prediction. Without task-specific fine-tuning, this framework achieves competitive zero-shot performance on eight unseen real-world datasets and scales up to 700 distinct event types.

## Synthesis and Verification of Register Automata over Data Sequences (Technical Report)

DevFeed: [Synthesis and Verification of Register Automata over Data Sequences (Technical Report)](<https://devfeed.tech/articles/synthesis-and-verification-of-register-automata-over-data-sequences-technical-report-94682.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2511.19100>)

Author: Chih-Duo Hong, Hongjian Jiang, Anthony W. Lin, Oliver Markgraf, Julian Parsert, Tony Tan

Published: 2026-10-22T04:00:00Z

Content type: article

Language: en

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

Topics: [Sequences](<https://devfeed.tech/topics/sequences.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [concept-learning](<https://devfeed.tech/topics/concept-learning.md>), [dfa](<https://devfeed.tech/topics/dfa.md>), [polynomials](<https://devfeed.tech/topics/polynomials.md>)

Tags: [alignment](<https://devfeed.tech/tags/alignment.md>), [alphabet](<https://devfeed.tech/tags/alphabet.md>), [application](<https://devfeed.tech/tags/application.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [classifiers](<https://devfeed.tech/tags/classifiers.md>), [comparisons](<https://devfeed.tech/tags/comparisons.md>), [control](<https://devfeed.tech/tags/control.md>), [costs](<https://devfeed.tech/tags/costs.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-fl](<https://devfeed.tech/tags/cs-fl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [dra](<https://devfeed.tech/tags/dra.md>), [fixed](<https://devfeed.tech/tags/fixed.md>), [framework](<https://devfeed.tech/tags/framework.md>), [guarantees](<https://devfeed.tech/tags/guarantees.md>), [implement](<https://devfeed.tech/tags/implement.md>), [infinite](<https://devfeed.tech/tags/infinite.md>), [languages](<https://devfeed.tech/tags/languages.md>), [large](<https://devfeed.tech/tags/large.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [synthesis](<https://devfeed.tech/tags/synthesis.md>), [verification](<https://devfeed.tech/tags/verification.md>)

### AI overview

The report presents a method to synthesize Deterministic Register Automata (DRAs) and check their robustness on ordered rational data sequences. Its exact checker reduces robustness to coverability and then shortest paths; for a fixed register count it runs in polynomial time, with exponential dependence on that count. It can find variable-length witnesses, including for alignment costs such as DTW, without a preset length bound. The authors also use learned DRAs as interpretable surrogates for neural sequence classifiers: a PAC validation loop refines rejected candidates, can return genuine non-robustness witnesses, and accepts surrogates with statistical guarantees for fidelity and local stability. Their RAExc implementation is evaluated on benchmark languages and sequentialized MNIST; the abstract reports high accuracy and verified robustness of learned automata.

### Source excerpt

arXiv:2511.19100v2 Announce Type: replace-cross Abstract: Data sequences, such as time series, traces, and sensor streams, are words over large or infinite data domains. Their properties often depend on comparisons between values at different positions, which finite automata cannot express directly. We study synthesis and verification for Deterministic Register Automata (DRAs) over ordered data sequences. DRAs combine finite control with registers for remembered data values, providing compact symbolic descriptions of quantitative sequence languages. Our technical core is a local-search synthesis procedure and an exact robustness checker for DRAs. We fix the alphabet theory to ordered rationals $(\Q;<,=)$ with rational guard constants and use non-negative affine accumulator costs. For a fixed number of DRA registers, the checker runs in polynomial time; its dependence on that number is exponential. It reduces robustness to coverability and then to a shortest-path problem. Since the search ranges over $\Q^+$, it produces variable-length witnesses and, for RAA-definable alignment costs such as DTW, explores a counterexample space with no a priori length bound. As an application, we use learned DRAs as interpretable surrogates for queried neural sequence classifiers. A probably approximately correct (PAC) validation loop refines rejected candidates, returns genuine non-robustness witnesses for the classifier when found, and accepts a surrogate only with statistical guarantees for fidelity and local stability. We implement the framework in $\textsc{RAExc}$, enabling the synthesis of interpretable automata and the verification of their robustness. We evaluate the approach on benchmark languages and sequentialized MNIST, demonstrating high accuracy while verifying the robustness of the learned automata.

## Building LLM Primitives from Scratch in TypeScript: Tokenizers, Vectors, and Attention

DevFeed: [Building LLM Primitives from Scratch in TypeScript: Tokenizers, Vectors, and Attention](<https://devfeed.tech/articles/building-llm-primitives-from-scratch-in-typescript-tokenizers-vectors-and-attention-61080.md>)

Original publisher: [Read original article](<https://www.sitepoint.com/building-llm-primitives-from-scratch-typescript/>)

Author: SitePoint Team

Published: 2026-09-28T14:42:29Z

Content type: tutorial

Language: en

Sources: [SitePoint](<https://devfeed.tech/sources/sitepoint.md>)

Topics: [TypeScript](<https://devfeed.tech/topics/typescript.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [vector embeddings in AI](<https://devfeed.tech/topics/vector-embeddings-in-ai.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [cosine-similarity](<https://devfeed.tech/tags/cosine-similarity.md>), [javascript](<https://devfeed.tech/tags/javascript.md>), [llm](<https://devfeed.tech/tags/llm.md>), [programming](<https://devfeed.tech/tags/programming.md>), [tokenization](<https://devfeed.tech/tags/tokenization.md>), [tokenizers](<https://devfeed.tech/tags/tokenizers.md>), [typescript](<https://devfeed.tech/tags/typescript.md>), [vectors](<https://devfeed.tech/tags/vectors.md>)

### AI overview

This tutorial builds three LLM-related primitives from scratch in TypeScript: a byte-pair encoding tokenizer, cosine-similarity vector search, and scaled dot-product attention. It explains their roles and limits, emphasizing that these components are not a complete or production LLM.

### Source excerpt

You will understand and implement the foundational mechanics of LLMs--including BPE tokenization, vector search, and self-attention math--entirely in zero-dependency TypeScript. Continue reading Building LLM Primitives from Scratch in TypeScript: Tokenizers, Vectors, and Attention on SitePoint.

## A Google ADK Pattern for Safe Parallel Agent Workflows

DevFeed: [A Google ADK Pattern for Safe Parallel Agent Workflows](<https://devfeed.tech/articles/two-agents-made-the-right-call-and-still-broke-the-workflow-59360.md>)

Original publisher: [Read original article](<https://hackernoon.com/two-agents-made-the-right-call-and-still-broke-the-workflow?source=rss>)

Author: Raju Dandigam

Published: 2026-09-24T06:17:50Z

Content type: tutorial

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [google-adk](<https://devfeed.tech/topics/google-adk.md>), [Parallelism](<https://devfeed.tech/topics/parallelism.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Graphs](<https://devfeed.tech/topics/graphs.md>), [Google](<https://devfeed.tech/topics/google.md>), [execution](<https://devfeed.tech/topics/execution.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [agent-workflows](<https://devfeed.tech/tags/agent-workflows.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-testing](<https://devfeed.tech/tags/ai-testing.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [concurrently](<https://devfeed.tech/tags/concurrently.md>), [distributed-systems](<https://devfeed.tech/tags/distributed-systems.md>), [google](<https://devfeed.tech/tags/google.md>), [google-adk](<https://devfeed.tech/tags/google-adk.md>), [multi-agent-systems](<https://devfeed.tech/tags/multi-agent-systems.md>), [software-architecture](<https://devfeed.tech/tags/software-architecture.md>), [tests](<https://devfeed.tech/tags/tests.md>), [topology](<https://devfeed.tech/tags/topology.md>), [typescript](<https://devfeed.tech/tags/typescript.md>)

### AI overview

This practical tutorial explains when multiple agents can safely run in parallel and when shared state or side effects require sequential execution. It presents Google ADK 2.0 graph-based workflows with explicit branches, joins, single-writer state, failure policies, and topology-based tests.

### Source excerpt

A practical Google ADK pattern for safe fan-out, joins, single-writer state, failure policy, and topology-based agent tests.

## Anthropic says Claude AI helped discover novel enzyme system

DevFeed: [Anthropic says Claude AI helped discover novel enzyme system](<https://devfeed.tech/articles/anthropic-says-claude-ai-helped-discover-novel-enzyme-system-59049.md>)

Original publisher: [Read original article](<https://indianexpress.com/article/technology/science/anthropic-says-claude-ai-helped-discover-novel-enzyme-system-10891634/>)

Author: Reuters

Published: 2026-09-24T03:11:14Z

Content type: news

Language: en

Sources: [Technology | The Indian Express](<https://devfeed.tech/sources/technology-the-indian-express.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [anthropic](<https://devfeed.tech/topics/anthropic.md>), [Claude](<https://devfeed.tech/topics/claude.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-discovered-gene-editing-system](<https://devfeed.tech/tags/ai-discovered-gene-editing-system.md>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [anthropic-claude-novel-enzyme-discovery](<https://devfeed.tech/tags/anthropic-claude-novel-enzyme-discovery.md>), [anthropic-life-sciences-research](<https://devfeed.tech/tags/anthropic-life-sciences-research.md>), [anthropic-wet-lab-san-francisco](<https://devfeed.tech/tags/anthropic-wet-lab-san-francisco.md>), [array-associated-reverse-transcriptases-art](<https://devfeed.tech/tags/array-associated-reverse-transcriptases-art.md>), [biology](<https://devfeed.tech/tags/biology.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-art-reverse-transcriptase-crispr](<https://devfeed.tech/tags/claude-art-reverse-transcriptase-crispr.md>), [dna](<https://devfeed.tech/tags/dna.md>), [life-sciences](<https://devfeed.tech/tags/life-sciences.md>), [model](<https://devfeed.tech/tags/model.md>), [patterns](<https://devfeed.tech/tags/patterns.md>), [research](<https://devfeed.tech/tags/research.md>), [science](<https://devfeed.tech/tags/science.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [technology](<https://devfeed.tech/tags/technology.md>), [technology-science](<https://devfeed.tech/tags/technology-science.md>)

### AI overview

Anthropic says its Claude model helped identify a novel enzyme system called array-associated reverse transcriptases, or ART, by analyzing large DNA databases. The system includes repeating DNA sequences resembling CRISPR patterns and an additional protein of unknown function.

### Source excerpt

Anthropic has named the system array-associated reverse transcriptases, or ART.

## Kodebits Day 95: Lazy Sequences \[FREE\]

DevFeed: [Kodebits Day 95: Lazy Sequences \[FREE\]](<https://devfeed.tech/articles/kodebits-day-95-lazy-sequences-free-55383.md>)

Original publisher: [Read original article](<https://www.kodeco.com/52032942-kodebits-day-95-lazy-sequences>)

Published: 2026-09-19T17:00:12Z

Content type: tutorial

Language: en

Sources: [Kodeco](<https://devfeed.tech/sources/kodeco-high-quality-programming-tutorials-ios-android-swift-kotlin-unity-and-more.md>)

Topics: [Sequences](<https://devfeed.tech/topics/sequences.md>), [Kotlin](<https://devfeed.tech/topics/kotlin.md>)

Tags: [article](<https://devfeed.tech/tags/article.md>), [free](<https://devfeed.tech/tags/free.md>), [kotlin](<https://devfeed.tech/tags/kotlin.md>), [sequences](<https://devfeed.tech/tags/sequences.md>)

### AI overview

A short Kotlin challenge for practicing sequences.

### Source excerpt

Practice sequences with a short kotlin challenge.

## Chaining Activities -- from text to vectors

DevFeed: [Chaining Activities -- from text to vectors](<https://devfeed.tech/articles/chaining-activities-from-text-to-vectors-35756.md>)

Original publisher: [Read original article](<https://temporal.io/blog/chaining-activities-from-text-to-vectors>)

Author: Houman Kargaran

Published: 2026-08-12T00:00:00Z

Content type: tutorial

Language: en

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

Topics: [data](<https://devfeed.tech/topics/data.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [on-prem](<https://devfeed.tech/topics/on-prem.md>), [pii](<https://devfeed.tech/topics/pii.md>), [Redis](<https://devfeed.tech/topics/redis.md>), [retry](<https://devfeed.tech/topics/retry.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [article](<https://devfeed.tech/tags/article.md>), [community](<https://devfeed.tech/tags/community.md>), [data-pipeline](<https://devfeed.tech/tags/data-pipeline.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [on-prem](<https://devfeed.tech/tags/on-prem.md>), [pii](<https://devfeed.tech/tags/pii.md>), [redis](<https://devfeed.tech/tags/redis.md>), [retries](<https://devfeed.tech/tags/retries.md>), [storage](<https://devfeed.tech/tags/storage.md>)

### AI overview

This guest post explains a Temporal workflow that chains discrete Activities to process complaints while maintaining a PII boundary. It uses an on-premises all-MiniLM-L6-v2 model for embeddings, caches vectors in Redis, stores them externally, and relies on deterministic workflow sequencing and retry policies.

### Source excerpt

Learn how to build a durable, PII-conscious data pipeline with Temporal using on-prem embeddings, idempotent Activities, external vector storage, and retries.

## Personalizing Airbnb search by learning from the guest journey

DevFeed: [Personalizing Airbnb search by learning from the guest journey](<https://devfeed.tech/articles/personalizing-airbnb-search-by-learning-from-the-guest-journey-56657.md>)

Original publisher: [Read original article](<https://airbnb.tech/ai-ml/personalizing-airbnb-search-by-learning-from-the-guest-journey/>)

Author: laurenmackevich

Published: 2026-07-27T19:00:17Z

Content type: article

Language: en

Sources: [Airbnb Engineering & Data Science](<https://devfeed.tech/sources/airbnb-engineering-data-science.md>)

Topics: [recommendation systems](<https://devfeed.tech/topics/recommendation-systems.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>), [Transformer](<https://devfeed.tech/topics/transformer.md>), [Model Development](<https://devfeed.tech/topics/model-development.md>)

Tags: [ai-ml](<https://devfeed.tech/tags/ai-ml.md>), [optimizations](<https://devfeed.tech/tags/optimizations.md>), [ranking](<https://devfeed.tech/tags/ranking.md>), [sequence](<https://devfeed.tech/tags/sequence.md>), [transformer](<https://devfeed.tech/tags/transformer.md>)

### AI overview

Airbnb describes a Transformer-based sequence modeling system that learns from years of guest behavior, including views, bookings, reviews, and cancellations, to improve personalized search ranking. The article explains challenges involving long, noisy event sequences, sparse booking signals, and the cost of training at scale.

### Source excerpt

How we built a Transformer-based sequence model that encodes years of guest behavior to surface the right listings at the right time.

## More whimsical OEIS sequences

DevFeed: [More whimsical OEIS sequences](<https://devfeed.tech/articles/more-whimsical-oeis-sequences-40522.md>)

Original publisher: [Read original article](<https://www.jeremykun.com/shortform/2026-05-22-1528/>)

Published: 2026-05-22T22:28:34Z

Content type: opinion

Language: en

Sources: [Jeremy Kun](<https://devfeed.tech/sources/jeremy-kun.md>)

Topics: [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [depths-of-oeis](<https://devfeed.tech/tags/depths-of-oeis.md>), [doom](<https://devfeed.tech/tags/doom.md>), [oeis](<https://devfeed.tech/tags/oeis.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [shortform](<https://devfeed.tech/tags/shortform.md>), [xkcd](<https://devfeed.tech/tags/xkcd.md>)

### AI overview

An informal survey of whimsical OEIS sequences, including sequences inspired by XKCD, non-reduced fractions, hexadecimal patterns, James Bond primes, random tables, and references to the number 666.

### Source excerpt

Here are some more whimsical OEIS sequences I came across. XKCD 2016 joked that "OEIS keeps rejecting my submissions," including one that gives "Integers in increasing order of width when printed in Helvetica." Well, two days after that comic was published (2018-07-09), Hugo Pfoertner published A316600, with a very precise definition. Then he did Arial. Randall Munroe missed a huge opportunity to commit to his bit and actually try to submit some of his sequences before publishing the comic.

## Unusual uses of OEIS sequences on GitHub

DevFeed: [Unusual uses of OEIS sequences on GitHub](<https://devfeed.tech/articles/unusual-uses-of-oeis-sequences-on-github-40521.md>)

Original publisher: [Read original article](<https://www.jeremykun.com/shortform/2026-04-13-0700/>)

Published: 2026-04-13T14:00:00Z

Content type: opinion

Language: en

Sources: [Jeremy Kun](<https://devfeed.tech/sources/jeremy-kun.md>)

Topics: [Sequences](<https://devfeed.tech/topics/sequences.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [GitHub](<https://devfeed.tech/topics/github.md>), [Algorithms](<https://devfeed.tech/topics/algorithms.md>)

Tags: [algorithms](<https://devfeed.tech/tags/algorithms.md>), [code](<https://devfeed.tech/tags/code.md>), [coding](<https://devfeed.tech/tags/coding.md>), [depths-of-oeis](<https://devfeed.tech/tags/depths-of-oeis.md>), [github](<https://devfeed.tech/tags/github.md>), [oeis](<https://devfeed.tech/tags/oeis.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [shortform](<https://devfeed.tech/tags/shortform.md>)

### AI overview

The article surveys unusual uses of OEIS sequences in open-source software on GitHub. It describes their use in the Mercury and Ziffers live-coding music frameworks, the Plato e-reader's pen-size options, and the GC Wizard geocaching app.

### Source excerpt

I went hunting for references to the OEIS in open source code, and found some weird ones. There are not one, but two live-coding music frameworks that use OEIS sequences as a source for "anything that can be sequenced" in music. I'm guessing that's used for choosing pseudorandom melodies, interesting rhythyms, or how to overlap tracks in different ways. The first project is called mercury, which is advertised as having "an extensive library of algorithms to generate or transform numbersequences that can modulate parameters.

## The OEIS meta sequence and subway stations

DevFeed: [The OEIS meta sequence and subway stations](<https://devfeed.tech/articles/the-oeis-meta-sequence-and-subway-stations-40520.md>)

Original publisher: [Read original article](<https://www.jeremykun.com/shortform/2026-04-09-0556/>)

Published: 2026-04-09T13:55:17Z

Content type: opinion

Language: en

Sources: [Jeremy Kun](<https://devfeed.tech/sources/jeremy-kun.md>)

Topics: [Sequences](<https://devfeed.tech/topics/sequences.md>), [Mathematics](<https://devfeed.tech/topics/mathematics.md>), [Database](<https://devfeed.tech/topics/database.md>)

Tags: [database](<https://devfeed.tech/tags/database.md>), [depths-of-oeis](<https://devfeed.tech/tags/depths-of-oeis.md>), [new-york-city](<https://devfeed.tech/tags/new-york-city.md>), [numberphile](<https://devfeed.tech/tags/numberphile.md>), [oeis](<https://devfeed.tech/tags/oeis.md>), [quirk](<https://devfeed.tech/tags/quirk.md>), [sequence](<https://devfeed.tech/tags/sequence.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [shortform](<https://devfeed.tech/tags/shortform.md>), [train](<https://devfeed.tech/tags/train.md>)

### AI overview

This commentary examines OEIS sequence A051070, whose nth term is the nth entry of sequence A_n, or -1 when that sequence lacks enough terms. It highlights unusually large or unknown values, subway-stop sequences included in the OEIS, and self-referential questions involving A051070 and A102288.

### Source excerpt

A051070 is a sequence about OEIS sequences. a(n) is the n-th term in sequence A_n (or -1 if A_n doesn't have enough terms). So the first term in A051070 is 1 because A000001 is the number of groups of order n, and that sequence has 1 as its entry in index 1. A000002 is the Kolakoski sequence (what? For another time) and has value 2 in entry 2. The sequence continues: 1, 2, 1, 0, 2, 3, 0, 7, 8, 4, 63, 1, 316, ...

## Waiting for PostgreSQL 19 - Sequence synchronization in logical replication.

DevFeed: [Waiting for PostgreSQL 19 - Sequence synchronization in logical replication.](<https://devfeed.tech/articles/waiting-for-postgresql-19-sequence-synchronization-in-logical-replication-33669.md>)

Original publisher: [Read original article](<https://www.depesz.com/2025/11/11/waiting-for-postgresql-19-sequence-synchronization-in-logical-replication/>)

Author: depesz

Published: 2025-11-11T12:24:23Z

Content type: tutorial

Language: en

Sources: [select \* from depesz;](<https://devfeed.tech/sources/select-from-depesz.md>)

Topics: [PostgreSQL](<https://devfeed.tech/topics/postgresql.md>), [Replication](<https://devfeed.tech/topics/replication.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [logical](<https://devfeed.tech/tags/logical.md>), [pg19](<https://devfeed.tech/tags/pg19.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [psql](<https://devfeed.tech/tags/psql.md>), [publications](<https://devfeed.tech/tags/publications.md>), [replication](<https://devfeed.tech/tags/replication.md>), [sequence](<https://devfeed.tech/tags/sequence.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [synchronization](<https://devfeed.tech/tags/synchronization.md>), [uncategorized](<https://devfeed.tech/tags/uncategorized.md>), [upgrade](<https://devfeed.tech/tags/upgrade.md>), [upgrades](<https://devfeed.tech/tags/upgrades.md>), [waiting](<https://devfeed.tech/tags/waiting.md>)

### AI overview

This article demonstrates PostgreSQL 19's planned sequence synchronization for logical replication. It sets up source and destination databases, tests sequence synchronization, finds that resynchronization must be manually invoked, and reports that synchronizing 10,000 sequences took about one second in the author's test.

### Source excerpt

First, on 9th of October 2025, Amit Kapila committed patch: Add "ALL SEQUENCES" support to publications. This patch adds support for the ALL SEQUENCES clause in publications, enabling synchronization/replication of all sequences that is useful for upgrades. Publications can now include all sequences via FOR ALL SEQUENCES. psql enhancements: \d shows publications for ... Continue reading "Waiting for PostgreSQL 19 - Sequence synchronization in logical replication."

## Formalising external APIs

DevFeed: [Formalising external APIs](<https://devfeed.tech/articles/formalising-external-apis-51504.md>)

Original publisher: [Read original article](<https://gleam.run/news/formalising-external-apis>)

Author: Louis Pilfold

Published: 2025-10-19T00:00:00Z

Content type: release

Language: en

Sources: [Gleam](<https://devfeed.tech/sources/gleam.md>)

Topics: [Tool](<https://devfeed.tech/topics/tool.md>), [JavaScript](<https://devfeed.tech/topics/javascript.md>), [conditionals](<https://devfeed.tech/topics/conditionals.md>), [adapter pattern](<https://devfeed.tech/topics/adapter-pattern.md>), [Erlang](<https://devfeed.tech/topics/erlang.md>), [code beautification](<https://devfeed.tech/topics/code-beautification.md>), [import](<https://devfeed.tech/topics/import.md>), [modules](<https://devfeed.tech/topics/modules.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>), [Functional programming](<https://devfeed.tech/topics/functional-programming.md>), [identifier](<https://devfeed.tech/topics/identifier.md>), [Exception handling](<https://devfeed.tech/topics/exception-handling.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [apis](<https://devfeed.tech/tags/apis.md>), [beam](<https://devfeed.tech/tags/beam.md>), [data](<https://devfeed.tech/tags/data.md>), [data-types](<https://devfeed.tech/tags/data-types.md>), [ecosystem](<https://devfeed.tech/tags/ecosystem.md>), [ecosystems](<https://devfeed.tech/tags/ecosystems.md>), [elixir](<https://devfeed.tech/tags/elixir.md>), [erlang](<https://devfeed.tech/tags/erlang.md>), [external](<https://devfeed.tech/tags/external.md>), [interface](<https://devfeed.tech/tags/interface.md>), [javascript](<https://devfeed.tech/tags/javascript.md>), [overhead](<https://devfeed.tech/tags/overhead.md>), [performance](<https://devfeed.tech/tags/performance.md>), [wrapper](<https://devfeed.tech/tags/wrapper.md>)

### AI overview

Gleam v1.13.0 adds a documented guide for externals and a dedicated JavaScript API for constructing and inspecting Gleam data. Existing JavaScript externals will need to migrate, with tooling and team help planned. The compiler also prunes redundant bit-array checks and warns about impossible bit patterns, unused recursive arguments, and inefficient list-length emptiness checks. Build output now reports dependency changes to support auditing locked versions. Many language-server improvements add or refine code actions, including extracting functions, expanding and collapsing pattern matches, and generating functions across modules; extraction uses a placeholder name because of an LSP limitation. Other changes improve type names in generated documentation and editor actions, syntax diagnostics, and formatter behavior.

### Source excerpt

Gleam is a type-safe and scalable language for the Erlang virtual machine and JavaScript runtimes. Today Gleam v1.13.0 has been published. Let's go over all the highlights now. External API for Gleam data One of Gleam's strengths is that it is part of the BEAM and JavaScript ecosystems, enabling Gleam programs to take advantage of code written in Erlang, Elixir, JavaScript, and more. This is a large part of how Gleam was able to become a practical production-ready language so quickly, by not restricting Gleam programmers to just the comparatively young Gleam package ecosystem. A function written in one of these other languages can be imported into a Gleam module as an "external function", and then called without any additional performance overhead. @external(erlang, "moon_base", "launch_spaceship") pub fn lift_off(countdown: Int) -> Result(Spaceship, LaunchError) Here the Erlang function launch_spaceship from the module moon_base is being imported. One restriction to external functions is that they will need to return data types that Gleam can understand. Gleam's "external type" feature lets the programmer refer to types defined in other languages, but if the programmer wants to be able to directly construct and pattern match data from external functions it will need to be in a format compatible with Gleam data types. Often a programmer will write a small wrapper function in the external language to convert the data, and also to make the interface adhere to Gleam patterns and conventions. -module(moon_base). -export([launch_spaceship/1]). launch_spaceship(Countdown) -> try Spaceship = launch_control:launch_spaceship(Countdown), {ok, Spaceship} catch error:no_fuel -> {error, no_fuel}; error:bad_weather -> {error, bad_weather} end. This launch_spaceship Erlang function wraps the function from the launch_control module, converting the exception-based API into a Result type, the Erlang representation of Gleam's Ok and Error variants. One thing that a Gleam programmer ma

## Libghostty Is Coming

DevFeed: [Libghostty Is Coming](<https://devfeed.tech/articles/libghostty-is-coming-51443.md>)

Original publisher: [Read original article](<https://mitchellh.com/writing/libghostty-is-coming>)

Published: 2025-09-22T00:00:00Z

Content type: release

Language: en

Sources: [Mitchell Hashimoto](<https://devfeed.tech/sources/mitchell-hashimoto.md>)

Topics: [Terminal](<https://devfeed.tech/topics/terminal.md>), [Library](<https://devfeed.tech/topics/library.md>), [Parsing](<https://devfeed.tech/topics/parsing.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>), [cross-platform](<https://devfeed.tech/topics/cross-platform.md>), [API](<https://devfeed.tech/topics/api.md>), [Zig](<https://devfeed.tech/topics/zig.md>), [C](<https://devfeed.tech/topics/c.md>), [GitHub Actions](<https://devfeed.tech/topics/github-actions.md>), [Visual Studio Code](<https://devfeed.tech/topics/visual-studio-code.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [c](<https://devfeed.tech/tags/c.md>), [cross-platform](<https://devfeed.tech/tags/cross-platform.md>), [d](<https://devfeed.tech/tags/d.md>), [github](<https://devfeed.tech/tags/github.md>), [github-actions](<https://devfeed.tech/tags/github-actions.md>), [hosting](<https://devfeed.tech/tags/hosting.md>), [library](<https://devfeed.tech/tags/library.md>), [parsing](<https://devfeed.tech/tags/parsing.md>), [terminal](<https://devfeed.tech/tags/terminal.md>)

### AI overview

Mitchell Hashimoto announces that libghostty, an embeddable library extracted from Ghostty's terminal-emulation core, is beginning to take shape. The first planned component is a zero-dependency library with an API for parsing terminal sequences and maintaining terminal state. A Zig API is available for early testing, while the C API is still forthcoming.

### Source excerpt

Libghostty Is Coming Over two years ago, in one of my first public talks about Ghostty, I shared my vision for : an embeddable library for any application to embed their own fully functional, modern, and fast terminal emulator. Libghostty is finally starting to take shape, and I'm excited to share more details about my plans for it.

[Next page](<https://devfeed.tech/topics/sequences.md?cursor=WyIyMDI1LTA5LTIyVDAwOjAwOjAwKzAwOjAwIiwgIjhlMmRhY2MyLTA5MDQtNGQ4OC04YmYzLTRiMTI2NDMwZWE4NSJd>)