CompanyAnthropic8 recent entries30 Jul 2026Identifying Implicit Bias in LLM-based Chat AI Toward People with Intellectual DisabilitiesarXiv:2607.26062v1 Announce Type: cross Abstract: Background: This work investigates the presence of implicit bias in Large Language Model (LLM)-based chat AI models directed toward people with intell→30 Jul 2026AdaMARP: An Adaptive Multi-Agent Interaction Framework for General Immersive Role-PlayingarXiv:2601.11007v2 Announce Type: replace-cross Abstract: LLM role-playing aims to portray arbitrary characters in interactive narratives, yet existing systems often suffer from limited immersion and
CompanyOpenAI8 recent entries30 Jul 2026Identifying Implicit Bias in LLM-based Chat AI Toward People with Intellectual DisabilitiesarXiv:2607.26062v1 Announce Type: cross Abstract: Background: This work investigates the presence of implicit bias in Large Language Model (LLM)-based chat AI models directed toward people with intell→7 Aug 2026What Current AI Benchmarks Leave Unmeasured: Modality, Search, Citations, and Implications (for Safety Evaluations)arXiv:2608.06202v1 Announce Type: cross Abstract: Large language model (LLM) benchmark evaluations are routinely used to support claims about model safety, reliability, and deployment readiness. Yet m→26 Aug 2026Olapa-MCoT: Enhancing the Chinese Mathematical Reasoning Capability of LLMsarXiv:2312.17535v2 Announce Type: replace Abstract: In the past two years, the outstanding performance of ChatGPT in multilingual and multitasking has led to large language models (LLMs) attracting wi→27 Aug 2026A Primer on Computational Semantics for Artificial Intelligence SystemsarXiv:2608.25022v1 Announce Type: new Abstract: As people adopt transformer-based language models (e.g., ChatGPT and Gemini) for an increasing number of use-cases, it is important to know how such mod→1 Sept 2026Token Counts Are Not Model Lineage: A Frozen-Threshold Holdout Study of Black-Box LLM API FingerprintingarXiv:2608.29930v1 Announce Type: new Abstract: Black-box model attribution is increasingly relevant when large language models (LLMs) are served through relay and reseller APIs. A tempting low-cost s→1 Sept 2026Gurukul AI: An Interactive AI-Driven Educational Platform for Indian Education SystemarXiv:2608.28611v1 Announce Type: cross Abstract: Recent advances in large language models (LLMs) like ChatGPT and LLaMA have transformed AI-driven education, but these systems are predominantly train→10 Sept 2026Experimental Analysis of Productive Interaction Strategy with ChatGPT: User Study on Function and Project-level Code Generation TasksarXiv:2508.04125v2 Announce Type: replace-cross Abstract: The application of Large Language Models (LLMs) is growing in the productive completion of Software Engineering tasks. Yet, studies investigat→14 Sept 2026Harness or Model? Isolating the Harness Effect in Agentic Coding with a Contamination-Controlled Private SuitearXiv:2609.11987v1 Announce Type: cross Abstract: An agentic coding system couples a language model to a harness: the tools, prompts and control flow that turn a chat model into an autonomous software
CompanyGoogle8 recent entries11 Sept 2026Black-Box Membership Inference via Word-Level Probability EstimationarXiv:2609.10611v1 Announce Type: cross Abstract: Membership inference attacks (MIAs) have emerged as critical tools for auditing privacy risks in large language models (LLMs), aiming to determine whe→14 Sept 2026Harness or Model? Isolating the Harness Effect in Agentic Coding with a Contamination-Controlled Private SuitearXiv:2609.11987v1 Announce Type: cross Abstract: An agentic coding system couples a language model to a harness: the tools, prompts and control flow that turn a chat model into an autonomous software→15 Sept 2026Realtime-Venus: A full-duplex interaction system with asynchronous delegationarXiv:2609.13814v1 Announce Type: new Abstract: Natural interaction in digital and physical environments requires continuous perception and timely responses. Spoken dialogue relies on acoustic and lin→15 Sept 2026MARCUS: An agentic, multimodal vision-language model for cardiac diagnosis and managementarXiv:2603.22179v2 Announce Type: replace Abstract: Cardiovascular disease remains the leading cause of global mortality, with progress hindered by human interpretation of complex cardiac tests. Curre→7 Oct 2026Image Bitstream Fine-grained Understanding for Privacy-Friendly AIoTarXiv:2610.08414v1 Announce Type: new Abstract: Image Bitstream Fine-grained Understanding (IBFU) aims to directly perform fine-grained classification and semantic description generation from encoded →9 Oct 2026Deception by Omission: Language Models Knowingly Hide Their MistakesarXiv:2610.11351v1 Announce Type: new Abstract: Large language models (LLMs) increasingly act as agents with little human oversight, so potential mistakes they make can go unnoticed. Users then depend→12 Oct 2026APEX-Harm: Evaluating Agent Safety in Shared Professional WorkspacesarXiv:2610.13623v1 Announce Type: cross Abstract: Professional agents increasingly work through command-line interfaces and multiple tools over shared workspaces containing files, messages, and busine→16 Oct 2026A team of Gemini 4 Argon agents beats MinecraftarXiv:2610.22319v1 Announce Type: new Abstract: We placed four Gemini 4 Argon agents in a sandboxed Minecraft survival world and gave them one goal: defeat the Ender Dragon. Each agent controlled a pl
CompanyMeta8 recent entries30 Sept 2026Same Bytes, Different Authority: Reserved-Token Representations in Chat-Template Prompt InjectionarXiv:2609.35932v1 Announce Type: cross Abstract: Prompt injection against LLM agents becomes much stronger when the injected instruction is wrapped in the model's own chat template. A forged template→30 Sept 2026Knowing Is Not Choosing: What Explicit Verification Adds Beyond Generative PreferencearXiv:2609.33142v2 Announce Type: replace Abstract: Generating a correct answer does not mean that a language model will select it. We separate factual recall into three steps: generating a correct ca→5 Oct 2026Prompted to Discriminate: Generalizing Malicious-Input Probes in the WildarXiv:2610.02413v1 Announce Type: new Abstract: LLM agents increasingly rely on activation probes as runtime monitors for prompt injection, jailbreaks, and unsafe requests, reading the model's own hid→7 Oct 2026Principled Under Pressure: Post-Training Decides Whether LLMs Act on Their Own Moral JudgmentarXiv:2610.08670v1 Announce Type: cross Abstract: Language models increasingly act as agents. An agent that says an action is wrong and then takes it anyway is a different failure from one that does n→7 Oct 2026Fork-and-Flush: Escaping Idea Basins in Autoresearch AgentsarXiv:2610.07447v1 Announce Type: new Abstract: Autoresearch agents tackle open-ended problems by repeatedly proposing candidate solutions, evaluating them, and using feedback to guide subsequent expe→8 Oct 2026How Fragile Is On-Device Language Model Safety? Localizing Safety-Critical Parameters for Sparse Fault AnalysisarXiv:2610.09000v1 Announce Type: cross Abstract: As small language models (SLMs) are increasingly deployed on resource-constrained and on-device platforms, including as components of agentic systems,→12 Oct 2026Agreement Without Alignment: Label-Free Rubric Clarification Increases LLM-Judge Consensus but Not Agreement with HumansarXiv:2610.13303v1 Announce Type: cross Abstract: LLM judges are increasingly deployed as panels, and disagreement among panel members is often read as a symptom of an unclear rubric. We test a label-→14 Oct 2026GPU Goldmines: Tunable WebGPU Kernels for Edge LLM InferencearXiv:2610.17321v1 Announce Type: cross Abstract: As demand for LLM inference grows and consumer GPUs grow more capable, edge deployment is becoming increasingly attractive. WebGPU is compelling in th
CompanyMistral8 recent entries25 May 2026It's the humans, not the data: Geopolitical bias in LLMs originates in post-training, amplified by the language of the promptarXiv:2605.23825v1 Announce Type: cross Abstract: It has generally been assumed that geopolitical bias in language models originates from the training data used during the pre-training phase. We teste→3 Jun 2026Fully Automated Identification of Lexical Alignment and Preference-Stage Shifts in Large Language ModelsarXiv:2606.03165v1 Announce Type: cross Abstract: The language used by digital chat assistants such as ChatGPT can diverge from human expectations (misalignment). Research, mostly on Scientific Englis→4 Jun 2026Bypassing Prompt Guards in Production with Controlled-Release PromptingarXiv:2510.01529v3 Announce Type: replace Abstract: Ball et al. recently established that prompt filtering for AI alignment faces a fundamental barrier: under standard cryptographic assumptions, no fi→7 Jul 2026The Moving Target: A Longitudinal Audit of Trustworthiness Drift Across Twelve Checkpoints of Open-Source Chat LLMsarXiv:2607.02587v1 Announce Type: cross Abstract: Model cards quote trust-benchmark scores without recording when they were measured, and the same number is routinely carried across successive checkpo→29 Jul 2026Less Data, Better Alignment: Data-Centric Multi-Evaluator Agreement for Preference OptimizationarXiv:2607.25136v1 Announce Type: new Abstract: Research on preference optimization often varies the training objective while holding the data fixed. We instead ask whether a small, high-confidence se→30 Jul 2026Identifying Implicit Bias in LLM-based Chat AI Toward People with Intellectual DisabilitiesarXiv:2607.26062v1 Announce Type: cross Abstract: Background: This work investigates the presence of implicit bias in Large Language Model (LLM)-based chat AI models directed toward people with intell→16 Sept 2026Competence-Preserving Resume Perturbations Expose Presentation Sensitivity in LLM ScreeningarXiv:2609.16517v1 Announce Type: cross Abstract: Resume screeners must infer job-relevant competence from resumes whose presentation can vary substantially in wording, structure, stylistic polish, an→23 Sept 2026Block-Level Weight-Space Structure Persists Under Post-Training: An Empirical Study Across LLM FamiliesarXiv:2609.26147v1 Announce Type: new Abstract: Modern LLMs are deployed as families of post-trained variants (base, instruct, chat, code) derived from a shared set of pre-trained weights. We present
CompanyxAI2 recent entries4 Jun 2026Bypassing Prompt Guards in Production with Controlled-Release PromptingarXiv:2510.01529v3 Announce Type: replace Abstract: Ball et al. recently established that prompt filtering for AI alignment faces a fundamental barrier: under standard cryptographic assumptions, no fi→27 Jul 2026Opaque Epistemic Mediation: How LLM Deployment Configurations Shape the Validation of Pseudo-SciencearXiv:2607.22513v1 Announce Type: cross Abstract: Commercial large language models are increasingly used as knowledge references, yet their stance on contested scientific claims is neither stable nor
CompanyDeepSeek8 recent entries18 Aug 2026Language Models that Think, Chat BetterarXiv:2509.20357v2 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards (RLVR) trains language models to use long chain-of-thought reasoning (CoT) in domains like mathematic→19 Aug 2026CABLE: Extending the Reach of Memory Retrieval via Complementary Antecedent-Based Linking and ExpansionarXiv:2608.17911v1 Announce Type: new Abstract: As LLM agents operate across structured workflows and sessions, preserving long-term history does not ensure that later contexts can recover relevant ev→28 Aug 2026Meta-Learning Where to Allocate Experts: Task-Conditioned Layer-Wise Compression for MoEsarXiv:2608.26650v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models route each token to a subset of expert networks, increasing capacity while keeping per-token computation sparse. In many→1 Sept 2026Token Counts Are Not Model Lineage: A Frozen-Threshold Holdout Study of Black-Box LLM API FingerprintingarXiv:2608.29930v1 Announce Type: new Abstract: Black-box model attribution is increasingly relevant when large language models (LLMs) are served through relay and reseller APIs. A tempting low-cost s→2 Sept 2026PersianAnonymizer: Evaluating LLM-Labeled Training for Efficient NER-based Anonymization in PersianarXiv:2609.00958v1 Announce Type: new Abstract: We target practical anonymization of Persian customer chats by training a compact NER model from LLM-labeled supervision and selecting the best labeler →2 Sept 2026Making Prospective Memory SLM-Shaped: Typed Intention Stores for Small-Model AgentsarXiv:2609.01272v1 Announce Type: new Abstract: Prospective memory means carrying out a deferred intention at the right future cue while other work continues. Benchmarks now isolate it as an agent ski→14 Sept 2026Harness or Model? Isolating the Harness Effect in Agentic Coding with a Contamination-Controlled Private SuitearXiv:2609.11987v1 Announce Type: cross Abstract: An agentic coding system couples a language model to a harness: the tools, prompts and control flow that turn a chat model into an autonomous software→18 Sept 2026Sampling Reveals Style: Unsupervised, Training-Free Discovery of Prompt-Conditional Stylistic Axes in LLM ActivationsarXiv:2609.19150v1 Announce Type: new Abstract: Large language models (LLMs) encode rich stylistic structure in their hidden activations, but discovering which stylistic dimensions are salient for a g
CompanyNVIDIA5 recent entries24 Jul 2026Domyn-Small: A European 10B Reasoning Language ModelarXiv:2607.20448v1 Announce Type: new Abstract: We introduce Domyn-Small, a 10-billion-parameter open-weight reasoning language model released under the MIT license. Domyn-Small is the product of an i→31 Jul 2026Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context MemoryarXiv:2607.28263v1 Announce Type: new Abstract: Transformer depth is not used uniformly: lower and middle layers build semantic representations, while upper layers increasingly specialize them for pre→4 Aug 2026RADAR: Rubric-Aware Dependency and Redundancy Analysis for LLM-as-Judge EvaluationarXiv:2608.01810v1 Announce Type: new Abstract: Rubric-based LLM-as-judge pipelines often assume that evaluation criteria provide independent signals. In practice, however, criteria can be behaviorall→18 Sept 2026PrefixBench-H100: Characterizing Prefix Reuse and Time-to-First-Token in H100 LLM ServingarXiv:2609.19657v1 Announce Type: cross Abstract: Repeated prompt prefixes are increasingly common in LLM serving workloads, appearing in system prompts, templated retrieval-augmented generation pipel→7 Oct 2026Offline AI Modules: Voice-First Offline Architecture, Hardware Reference Stack, Quantization and BenchmarkingarXiv:2610.07026v1 Announce Type: new Abstract: The Offline AI Modules workstream enables practical, low-power, and community-accessible deployment of voice-first AI systems that operate fully offline