Clay weekly context brief for the Computer Science category (ISO week 2026-W47). 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. Right Answers, Inconsistent Evidence: Source-Level Counterfactual Auditing of Large Audio-Language Models Source: cs.SD (Sound) Link: https://arxiv.org/abs/2611.15463 A large audio-language model (LALM) can answer "is a cat present?" about a mixture correctly, and its confidence can drop when the cat is removed, yet this does not show that it can tell the cat's presence from its absence. 2. Near-Quadratic Time Multivariate Polynomial Interpolation Source: cs.SC (Symbolic Computation) Link: https://arxiv.org/abs/2610.30907 Let $\F$ be a finite field. 3. Are We Ready for Snow-tonomous Driving? Challenges Toward Robust Automated Driving in Extreme Weather Source: cs.RO (Robotics) Link: https://arxiv.org/abs/2611.15217 Automated driving has gained remarkable momentum over the past decade, with both industry and academia investing substantial resources in developing perception and control solutions to advance vehicle autonomy. 4. Coinduction in Grounded Arithmetic Source: cs.PL (Programming Languages) Link: https://arxiv.org/abs/2611.00176 A grounded arithmetic counts a sentence true only when a terminating process establishes it; a sentence no such process settles has no value. 5. Inference Validation for Regulated AI: A Governed Decision Pipeline for LLM Serving Source: cs.PF (Performance) Link: https://arxiv.org/abs/2611.12737 Deploying large language models in regulated industries requires more than picking the fastest inference engine: it requires evidence that a configuration meets contractual and legal obligations before it serves a single customer request. 6. Designing Institutions for Autonomous AI Source: cs.SI (Social and Information Networks) Link: https://arxiv.org/abs/2611.17114 As autonomous AI agents increasingly interact in persistent populations, their collective capabilities and alignment depend not only on individual models, but also on the institutions governing their interactions. 7. Planning-Stage Assessment of Battery-Supported Regional Islanding: A South Australian Case Study Source: cs.SY (Systems and Control) Link: https://arxiv.org/abs/2611.15364 Power systems with high shares of renewable generation may face increased operational vulnerability under regional islanding caused by the loss of major inter-area transmission interfaces. 8. KernelAudit: Contract-Aware Auditing of Validators for LLM-Generated GPU Kernels Source: cs.SE (Software Engineering) Link: https://arxiv.org/abs/2611.15506 The effectiveness of generating GPU kernels driven by a large language model (LLM) is highly dependent on the quality of the used validator (used to accept or reject a generated kernel candidate against the contracts of the given kernel-generation task), whose results inform reward assignment, candidate selection, and speedup claims during kernel generation. 9. Bit-Sparsity Aware Acceleration with Compact CSD Code on Generic Matrix Multiplication Source: cs.AR (Hardware Architecture) Link: https://arxiv.org/abs/2611.15136 The ever-increasing demand for matrix multiplication in artificial intelligence (AI) and generic computing emphasizes the necessity of efficient computing power accommodating both floating-point (FP) and quantized integer (QINT). 10. Enhancing Resolution and Reliability with Attention Mechanisms in Nuclear Accident Surrogate Modeling Source: cs.CE (Computational Engineering, Finance, and Science) Link: https://arxiv.org/abs/2611.14532 Severe accident (SA) analysis of nuclear power plants relies on system codes such as MAAP, whose computational cost limits large-scale applications including dynamic probabilistic safety assessment and reinforcement learning. 11. MITRA-Swarm: Injecting Explicit Linguistic Structure into Foundation Models for Classical Machine Translation Source: cs.CL (Computation and Language) Link: https://arxiv.org/abs/2611.14675 Classical Sanskrit poses three challenges unresolved by single-call monolithic models: morphophonemic fusion Sandhi, dense nominal compounding Sam\=asa, and discourse-level zero-anaphora forced by poetic meter. 12. Incrementally Verifiable Oracle Aided Computation Source: cs.CR (Cryptography and Security) Link: https://arxiv.org/abs/2611.14698 Incrementally verifiable computation (IVC) allows a prover to maintain a short proof that can be efficiently updated after each computation step, without retaining or recomputing the preceding execution. 13. Fast Slice-Wise Deep Learning Motion Correction for Spinal Cord Functional MRI Source: cs.CV (Computer Vision and Pattern Recognition) Link: https://arxiv.org/abs/2611.14677 Respiratory and bulk motion displace the spinal cord between the volumes of a functional MRI time series. 14. Probe Before You Preach: Conformal Ambiguity Sets and Neuro-Symbolic Counterfactual Probing for Diagnosing Algebraic Misconceptions in Middle School Mathematics Source: cs.CY (Computers and Society) Link: https://arxiv.org/abs/2611.14557 Distinct mathematical misconception rules can produce the same incorrect answer, making diagnosis from a single response ambiguous. 15. The Ultimate Karp-Lipton Collapse Source: cs.CC (Computational Complexity) Link: https://arxiv.org/abs/2611.16096 Karp and Lipton proved that if $NP\subseteq P/poly$, then $PH$ collapses. 16. Beyond Foreign-Key Connectivity: Semantic Relationship Representation for Relational Deep Learning Source: cs.DB (Databases) Link: https://arxiv.org/abs/2611.14706 Relational databases encode information through tables and primary--foreign key constraints. 17. Auditing AI peer reviewers: a benchmark on real scientific papers Source: cs.DL (Digital Libraries) Link: https://arxiv.org/abs/2611.17052 LLM-based peer review is now used at production scale, but its evaluation remains limited. 18. Explicit Subspace Evasive Sets via Subspace Designs Source: cs.DM (Discrete Mathematics) Link: https://arxiv.org/abs/2611.14909 Given a finite field $\mathbb F_q$, we say that a set $A\subseteq \mathbb F^n_q$ is an $(r,c)$ subspace evasive set if and only if for any affine subspace $U\subseteq \mathbb F^n_q$ with bounded dimension $\dim(U)\leq r$, the intersection size $|A\cap U|$ is at most $c$. 19. Source Assignment: Hardness, Approximation, and Parameterized Algorithms Source: cs.DS (Data Structures and Algorithms) Link: https://arxiv.org/abs/2611.14786 In Source Assignment, each pointer represents a substring of a common reference string and may have several possible source occurrences. 20. Al-Ghasali: A Computational Framework for Simulating Islamic Jurisprudential Rulings via Finite State Automata and Large Language Models Source: cs.FL (Formal Languages and Automata Theory) Link: https://arxiv.org/abs/2611.17235 This paper presents Al Ghasali, a computational framework designed to formalize and simulate Islamic jurisprudential (Fiqh) rulings. 21. Neural Render Proxies for Interactive and Differentiable Lighting Source: cs.GR (Graphics) Link: https://arxiv.org/abs/2611.17185 Within the CG animation production pipeline, the challenges artists tackle in Lighting can be immense. 22. Hear What You Eat: In-Ear Passive Acoustic Sensing for Multi-task Dietary Monitoring Source: cs.HC (Human-Computer Interaction) Link: https://arxiv.org/abs/2611.14878 The increasing obesity health concerns are driving the advancement of continuous dietary monitoring. 23. When Direction Matters: Diagnosing Relational Instruction Following in Dense Retrieval Source: cs.IR (Information Retrieval) Link: https://arxiv.org/abs/2611.15417 Dense retrievers increasingly claim the ability to condition retrieval on natural-language relevance criteria beyond a short query. 24. DecArena: Evaluating Large-Scale Decentralized Multi-Agent Collaboration Source: cs.MA (Multiagent Systems) Link: https://arxiv.org/abs/2611.15521 Emerging personal AI agents raise fundamental scalability and privacy concerns for centralized multi-agent systems (MASs), which rely on a global orchestrator with broad access to agents, resources, and communication. 25. An Asynchronous Federated Learning Platform for Lung Adenocarcinoma Grading via Foundation Models and Structural Pruning Source: cs.LG (Machine Learning) Link: https://arxiv.org/abs/2611.14590 Classification of lung adenocarcinoma into poorly differentiated (PD) and well-to-moderately differentiated (WMD) subtypes plays an important role in patient survival; however, consistent classification is difficult due to inter-observer variability among pathologists. 26. CineWild: Balancing Art and Robotics for Ethical Wildlife Documentary Filmmaking Source: cs.MM (Multimedia) Link: https://arxiv.org/abs/2509.24921 Unmanned aerial vehicles (UAVs) have become powerful tools across domains from industry to the arts. 27. Where Polanyi Meets Galois: Tacit Knowing as an Algebro-Deterministic Operation, Toward Re-rooting AI-Generated Science Source: cs.NE (Neural and Evolutionary Computing) Link: https://arxiv.org/abs/2610.16102 Tacit knowing is an operating regime of one memory substrate, not a separate kind of content. 28. Agentic Orchestration for SLA-Constrained Network Slicing in Programmable Mobile Data Planes Source: cs.NI (Networking and Internet Architecture) Link: https://arxiv.org/abs/2611.16801 The Third Generation Partnership Project (3GPP) Technical Specification 23.501 (Release 17) defines Maximum Bit Rate (MBR) enforcement at the User Plane Function (UPF) in fifth-generation (5G) networks. 29. BF16 Component-Product Emulation of FP32 and FP64 GEMM on Intel AMX Source: cs.MS (Mathematical Software) Link: https://arxiv.org/abs/2609.04663 Modern CPUs increasingly integrate high-throughput matrix engines optimized for low-precision AI workloads, while many scientific computing applications still rely on FP32 and FP64 GEMM to meet their numerical accuracy requirements. 30. BunshinRL: Whole-Workspace Forking for Agentic Reinforcement Learning Rollouts Source: cs.OS (Operating Systems) Link: https://arxiv.org/abs/2611.10453 Agentic reinforcement learning (RL) rollouts interleave model reasoning with tool calls in persistent workspaces. 31. DeALR: Decoupled Audio-Latent Optimization with Acoustic Backtracking for Test-Time Reasoning Source: cs.SD (Sound) Link: https://arxiv.org/abs/2611.15439 Large audio-language models (LALMs) typically encode audio once and conduct subsequent multi-step reasoning in a language-dominated space, weakening access to fine-grained acoustic evidence. 32. Program Learning with Verifiable Rewards: Distilling Expert Traces into Programs Source: cs.SC (Symbolic Computation) Link: https://arxiv.org/abs/2610.19148 Small language models struggle with tool-use tasks that require following multi-step procedures. 33. ARTS: An Open-Source, Reproducible Three-Arm Platform for Autonomy Research in Minimally Invasive Surgery Source: cs.RO (Robotics) Link: https://arxiv.org/abs/2611.15135 Research on surgical autonomy needs open platforms that any laboratory can build, yet the most widely shared platform depends on donated clinical hardware, and open alternatives on industrial arms each lack part of the surgical chain. Sources in this brief: cs.AR (Hardware Architecture); cs.CC (Computational Complexity); cs.CE (Computational Engineering, Finance, and Science); 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.DL (Digital Libraries); cs.DM (Discrete Mathematics); cs.DS (Data Structures and Algorithms); cs.FL (Formal Languages and Automata Theory); cs.GR (Graphics); cs.HC (Human-Computer Interaction); cs.IR (Information Retrieval); cs.LG (Machine Learning); cs.MA (Multiagent Systems); cs.MM (Multimedia); cs.MS (Mathematical Software); cs.NE (Neural and Evolutionary Computing); cs.NI (Networking and Internet Architecture); cs.OS (Operating Systems); cs.PF (Performance); cs.PL (Programming Languages); 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 33 of 470 available items for this weekly brief.