Clay weekly context brief for the Computer Science category (ISO week 2026-W39).

Clay tracks publications from the Computer Science feed list. Below are recent items from this category, each with its source and a short description of what the publication covers when one is available in the source feed.

Recent publications:
1. Reading Emotions in the Token Space: Discriminative Adaptation of SpeechLLMs for Emotion Recognition
   Source: cs.SD (Sound)
   Link: https://arxiv.org/abs/2609.20081
   SpeechLLMs have shown strong potential for emotion recognition, yet they read the predicted emotion off a generative decoder not suited for classification: it can emit labels outside the target set and favors frequent classes.
2. Views: A Hardware-Aware Recursively Labeled Graph Database Model for Knowledge Representation and Reasoning
   Source: cs.SC (Symbolic Computation)
   Link: https://arxiv.org/abs/2508.18123
   Knowledge representation remains a central challenge for reasoning-centered artificial intelligence, particularly when semantic structures involve relations over relations, contextual annotations, and recursively nested descriptions.
3. Winning a Won Game: Strict Reach-Avoid-Stay Control Barrier Functions for High-Dimensional Black-Box Systems
   Source: cs.RO (Robotics)
   Link: https://arxiv.org/abs/2609.19449
   Robots must complete their tasks and maintain the achieved outcomes while avoiding safety failures at all times.
4. CatchBench: When Can an Agent Failure Be Caught?
   Source: cs.PF (Performance)
   Link: https://arxiv.org/abs/2608.22808
   When can an agent failure be caught?
5. DeltaSelect: Affordable A/B Testing for Coding Agents
   Source: cs.SE (Software Engineering)
   Link: https://arxiv.org/abs/2609.19607
   Coding-agent benchmarks are built for broad and comprehensive comparisons, not frequent development decisions.
6. Improving Cross-Lingual Transfer for Sequential Sentence Classification in Research Papers via Structural Similarity
   Source: cs.DL (Digital Libraries)
   Link: https://arxiv.org/abs/2609.19650
   Sequential sentence classification (SSC) is an essential task for structuring scientific publications, and extending SSC research to languages other than English can improve accessibility to scientific knowledge in multilingual digital libraries.
7. Reachability, Not Observation: Containing Systems Whose Wiring Changes
   Source: cs.CR (Cryptography and Security)
   Link: https://arxiv.org/abs/2609.19720
   Containment decisions -- where to put a firewall, which links to monitor, what a program may reach -- are computed from an observed structure, and observation is a snapshot.
8. Pre-train to Gain: Robust Learning Without Clean Labels
   Source: cs.NE (Neural and Evolutionary Computing)
   Link: https://arxiv.org/abs/2511.20844
   Training deep networks with noisy labels leads to poor generalization and degraded accuracy due to overfitting to label noise.
9. FedVideoMAE: Efficient Federated Video Moderation with Differential Privacy and Secure Aggregation
   Source: cs.MM (Multimedia)
   Link: https://arxiv.org/abs/2512.18809
   Short-form video moderation is increasingly pushed toward edge and privacy-sensitive settings, where users may intend videos for a limited audience, such as friends or private groups, but sending raw clips to a central server can broaden exposure, consume bandwidth, and add moderation latency.
10. CoRe: Coherence and Relational Alignment for Multivariate Time Series Forecasting
   Source: cs.LG (Machine Learning)
   Link: https://arxiv.org/abs/2609.19670
   Direct forecasting has become a standard paradigm for multivariate time-series forecasting because it predicts the full future horizon in a single pass.
11. Reasoning Quality Matters: Combating Reasoning Collapse in LLM-based Embedding Learning
   Source: cs.IR (Information Retrieval)
   Link: https://arxiv.org/abs/2609.20563
   Large Language Models (LLMs) have recently shown strong potential for producing context-rich text embeddings for retrieval.
12. SplashSplat: Reconstructing Splashing Liquids from Real-World Multi-View Videos
   Source: cs.GR (Graphics)
   Link: https://arxiv.org/abs/2609.20818
   A splash lives for a fraction of a second: sheets tear into ligaments and droplets, appearance is view-dependent and nearly textureless, and little persists long enough to track.
13. Metric Weighted Edit Distance: $(3+\varepsilon)$-Approximation in $\widetilde O_\varepsilon(N^{1.6})$ Time
   Source: cs.DS (Data Structures and Algorithms)
   Link: https://arxiv.org/abs/2609.20796
   For every $0 < \varepsilon \le 1$, we give a randomized $(3+\varepsilon)$-approximation to weighted edit distance when the costs form a metric on the alphabet augmented with a gap symbol.
14. Brief Announcement: Fair Binding for Hidden-State Authorization in Byzantine SMR
   Source: cs.DC (Distributed, Parallel, and Cluster Computing)
   Link: https://arxiv.org/abs/2608.17349
   Validated Byzantine SMR assumes that replicas can evaluate the validity of an ordered command.
15. ZigZag Trie: A Novel Index for Contextual Queries
   Source: cs.DB (Databases)
   Link: https://arxiv.org/abs/2609.19914
   There is increasing interest in queries about the context of a string $P$ in a longer text $T$, i.e., the set of all string pairs $(L,R)$, with $|L|=|R|=q$, for a given $q$, such that the string $LPR$ occurs in $T$.
16. Before the Arrest: Benchmarking LLMs on Criminal Profiling from Incomplete Evidence
   Source: cs.CY (Computers and Society)
   Link: https://arxiv.org/abs/2609.19965
   Large Language Models (LLMs) are increasingly applied to legal and criminal justice tasks, yet existing work focuses almost exclusively on post-arrest scenarios where the suspect's identity is already known, leaving the critical pre-arrest challenge of inferring suspect characteristics from incomplete evidence largely unexplored.
17. Hamilton decompositions of equal-side directed tori
   Source: cs.DM (Discrete Mathematics)
   Link: https://arxiv.org/abs/2605.04734
   Let $D_d(m)$ be the Cartesian product of $d$ positively oriented directed cycles of length $m$.
18. Chain-of-Thought Entropy as a Reliability Signal: A Preregistered Reproduction
   Source: cs.CL (Computation and Language)
   Link: https://arxiv.org/abs/2609.19606
   This empirical study is an independent reproduction of the dissociation Zhao reported in 2026.
19. IMFD: End-to-end Multi-Face Forgery Detection through Instruction-based Large Vision-Language Models
   Source: cs.CV (Computer Vision and Pattern Recognition)
   Link: https://arxiv.org/abs/2609.19693
   The rapid increase of deepfakes has raised significant concerns due to their spread on social media.
20. ResNLS: An Improved Model for Stock Price Forecasting
   Source: cs.CE (Computational Engineering, Finance, and Science)
   Link: https://arxiv.org/abs/2312.01020
   Stock prices forecasting has always been a challenging task.
21. Legislating World-Model-Based Planning with Legal Reasoning
   Source: cs.LO (Logic in Computer Science)
   Link: https://arxiv.org/abs/2609.15113
   As robotic systems grow more general, legal norms are needed to integrate them into society.
22. S4R: Scaling for Rigid-Body Interpenetration Resolution
   Source: cs.CG (Computational Geometry)
   Link: https://arxiv.org/abs/2609.20524
   Rigid-body interpenetration frequently occurs in procedurally assembled and generated scenes and must be removed before downstream applications such as physical simulation.
23. Efficient Randomized Communication Without Large Monochromatic Rectangles
   Source: cs.CC (Computational Complexity)
   Link: https://arxiv.org/abs/2609.20763
   In this paper, we construct a total Boolean function with $\widetilde{O}(\log n)$ randomized communication protocol, while any monochromatic rectangle has density at most $O(2^{-\mathrm{poly}(n)})$.
24. New directions in dynamical expectation estimation
   Source: cs.ET (Emerging Technologies)
   Link: https://arxiv.org/abs/2609.18246
   Computing dynamical expectation values typically relies on approximations optimized for the state or observable separately, without accounting for how their errors combine in the final expression.
25. What Do Current Systematic Generalization Tasks Miss? A Reasoning-Centered Analysis
   Source: cs.AI (Artificial Intelligence)
   Link: https://arxiv.org/abs/2609.19212
   Systematic generalization, the ability to solve novel problems by recombining known atomic elements, is central to human intelligence but difficult to study rigorously under controlled settings.
26. Experienced partisan segregation across patterns of mobility behavior in US cities
   Source: cs.SI (Social and Information Networks)
   Link: https://arxiv.org/abs/2609.19285
   Residential and social sorting along political lines structure Americans' daily interactions, often limiting encounters with opposing viewpoints.
27. Optimal Control and Closed-Loop Stability of Droplet Transport in a Microchannel
   Source: cs.SY (Systems and Control)
   Link: https://arxiv.org/abs/2609.19320
   Understanding the efficient transport of fluid droplets in confined geometries has been a domain of interest for industrial applications in recent times.
28. Foreground Voice Activity Detection: Learning Speaker Selectivity from Supervision
   Source: cs.SD (Sound)
   Link: https://arxiv.org/abs/2609.19856
   Voice activity detection (VAD) fronts most voice-agent pipelines, yet production detectors treat all human speech, background talkers included, as valid activity; in crowded settings this floods recognition, stalls turn-taking, and triggers false barge-in.
29. On deciding transcendence of power series
   Source: cs.SC (Symbolic Computation)
   Link: https://arxiv.org/abs/2504.16697
   It is well known that algebraic power series are differentially finite (D-finite): they satisfy linear differential equations with polynomial coefficients.
30. From Wizard-of-Oz Human-Robot Dialogue Collection to a Taxonomy of Robot Response Decisions: A Retrospective Analysis of Assistive Pilot Interactions
   Source: cs.RO (Robotics)
   Link: https://arxiv.org/abs/2609.19447
   Robots that follow natural-language instructions in everyday indoor environments must act on incomplete human utterances.
31. Estimating the Energy Footprint of Software Systems: a Primer
   Source: cs.PF (Performance)
   Link: https://arxiv.org/abs/2407.11611
   In Green Software Development, quantifying the energy footprint of a software system is one of the most basic activities.

Sources in this brief: cs.AI (Artificial Intelligence); cs.CC (Computational Complexity); cs.CE (Computational Engineering, Finance, and Science); cs.CG (Computational Geometry); cs.CL (Computation and Language); cs.CR (Cryptography and Security); cs.CV (Computer Vision and Pattern Recognition); cs.CY (Computers and Society); cs.DB (Databases); cs.DC (Distributed, Parallel, and Cluster Computing); cs.DL (Digital Libraries); cs.DM (Discrete Mathematics); cs.DS (Data Structures and Algorithms); cs.ET (Emerging Technologies); cs.GR (Graphics); cs.IR (Information Retrieval); cs.LG (Machine Learning); cs.LO (Logic in Computer Science); cs.MM (Multimedia); cs.NE (Neural and Evolutionary Computing); cs.PF (Performance); cs.RO (Robotics); cs.SC (Symbolic Computation); cs.SD (Sound); cs.SE (Software Engineering); cs.SI (Social and Information Networks); cs.SY (Systems and Control).
Selected 31 of 362 available items for this weekly brief.