628 papers recorded in this category. Newest first.
- MDSKILL: Learning to Internalize Skill Memory through Multi-View Distillation for Efficient Dynamic Agent Self-Evolution
Skill memories encode past experiences as textual prompts to improve reinforcement learning (RL) for self-evolving agents. Complementing RL, On-Policy Self-Distillation (OPSD) prov…
2026-11-03
- ARLoop: Fully-Open Pre-trained Looped Transformers with Attention Residuals
Looped transformers increase effective depth by repeatedly applying a shared set of transformer layers, offering a parameter-efficient way to scale computation. However, standard r…
2026-11-03
- What Does the Model See? Representation Validity in Multimodal LLM-Based Video Analysis
Multimodal large language models (MLLMs) allow social scientists to analyze video at scale, but researchers must decide which parts of a video to provide as input. We use the term …
2026-11-03
- When Self-Verification Hurts: Calibrated Answer Repair for Financial Question Answering
Financial question answering can fail even when a model extracts a plausible number from the supplied evidence: the answer may attach it to the wrong entity, period, unit, or accou…
2026-11-03
- GM League Arena: An Interactive Long-Horizon Task Benchmark Based on Professional Sports League Management
Long-horizon decision-making under sustained competition requires large language models (LLMs) to balance immediate investment with future resource needs. However, local success in…
2026-11-03
- The Right Source Is Not Enough: Understanding History Interference in Speech-LLM Memory
Speech large language models retain audio history to support multi-turn conversations. Yet remembering an event does not ensure that its information is used only when relevant. We …
2026-11-03
- ToolSpace: Scaling Tool-Centric Environment Synthesis for Long-Horizon Agents
Large language models are deployed as long-horizon agents that compose diverse tools across MCP services into executable workflows. Existing MCP environment-scaling methods are dat…
2026-11-03
- Whose Effect Is It? How Objectives Inflate Learned Subspace Interventions
Interpretability often locates concepts by copying part of a language model's state from one input into another (activation patching). A larger output change counts as stronger evi…
2026-11-03
- Who is the Pronoun Pro? Comparing LLMs and Humans at Number-Disambiguation of 'they'
This study evaluates whether large language models (LLMs) can reliably replace or complement human research assistants in linguistically complex annotation. In a controlled annotat…
2026-11-03
- Shuffled OPSD: When Semantic Consistency Hurts On-Policy Self-Distillation?
Knowledge distillation improves a student model's reasoning ability through supervision from a teacher. On-policy self-distillation (OPSD) uses the same model as both teacher and s…
2026-11-03
- ufakzeka-kulak: Building and Evaluating a 31.6M-Parameter Turkish Speech Recognizer from Scratch
ufakzeka-kulak is an open Turkish speech-to-text model of 31.6M parameters, a FastConformer encoder with a token-and-duration transducer (TDT) and a CTC head, trained from scratch …
2026-11-02
- NitiBench+: A Temporal Thai Legal Question Answering Benchmark Across 17 Domains
Law changes over time. A court decides a case under the law in force on that day, and legal QA benchmarks grade answers against today's law. We checked 156,650 decisions of four Th…
2026-11-02
- VeriHarm-Bench: Measuring Harmful Success in LLM Coding Agents Verifiably
Large language model (LLM) coding agents increasingly operate with shell access, allowing them to write scripts, move files, alter permissions, and modify system configurations. Ho…
2026-11-02
- VIAS: Value-Isolating Activation Steering for Multi-Value Alignment and Compliance for Organisations
Organisations increasingly deploy large language model agents that must follow organisational values. We check whether activation steering, which adds vectors to a model's internal…
2026-11-02
- Online Label-Free Gradient Matching for Continual Language Model Training
Language models in customer-facing products must be continually updated, since the data they serve shifts over time. As retraining on the full accumulated data is impractical, each…
2026-11-02
- Optimize the Harness, Not the Model: Fitting a Minimal Agent to a Self-Hostable LLM for Investigative Analysis
Government intelligence teams need LLM agents that can analyze large, messy evidence collections, but security and cost constraints often rule out hosted frontier models and broad,…
2026-11-02
- IdeaTaste: Do LLMs Share the Research Taste of Human Researchers?
Research taste, the judgment of what to read, which problems to pursue, and which methods to try, increasingly shapes the work of AI scientists, yet it remains far less understood …
2026-11-02
- Same Outputs, Different Scores: Auditing the Unchecked Graders of Multimodal LLM Benchmarks
Every multimodal-LLM multiple-choice benchmark ends with a scoring rule that converts a model's response, often free-form text rather than a clean option letter, into a right/wrong…
2026-11-02
- Characterizing When Language Models Take Harmful Actions to Protect Their Assigned Goals
Recent language model (LM) safety failures have largely been studied as disparate phenomena. In agentic misalignment, a model blackmails an executive to avoid being replaced. In al…
2026-11-02
- Steering with Multilingual Lexicons for Sentiment Control in Low-Resource Languages
Activation steering offers a lightweight mechanism for controlling model behavior, but existing methods depend on sentence-level contrastive data that is unavailable for most of th…
2026-11-02
- MultiInpaintCOCO: Fine-Grained Visual Understanding in Multiple Languages
Previous research shows that vision-language (VL) models struggle to understand fine-grained image details and tend to focus on the overall layout. Moreover, any weakness in Englis…
2026-11-02
- LLMs Depend on Specialist Tools for Out-of-Distribution Generalization in MS/MS Structural Elucidation
A frontier language model was recently reported to solve 1D NMR structural elucidation without tools. We ask whether this capability transfers to tandem mass spectrometry (MS/MS) i…
2026-11-02
- They Know Where It Is, But Not What It Says: A Bilingual Hallucination Benchmark of Hadith Recall in Open LLMs
Large language models are increasingly consulted for religious knowledge, including the recall of specific Hadith (recorded sayings and actions of the Prophet Muhammad, peace be up…
2026-11-02
- From Language to Country to Valence: A Multilingual Audit of Stereotype Bias Across LLMs
Large language models (LLMs) are increasingly considered for education, recruitment, workplace decision support, and other consequential applications, making pre-deployment social-…
2026-11-02
- SMILE: A Structured Multi-Agent Clinical Reasoning for Liver Ultrasound Understanding
Automated liver ultrasound report generation improves the diagnosis of liver diseases by producing clinically structured descriptions of anatomical and disease-related findings. Ho…
2026-11-02
- Re$^3$Rank: Listwise Reranking with Reasoning-Conditioned Retrieval Heads
Attention-based rerankers score each candidate by the attention mass that retrieval heads route to it from the question, but the question, serving as a fixed query, carries limited…
2026-11-02
- FindLink: Learning to Assess Scientific Connections from Source-Grounded Premises
Scientific discovery from the literature requires assessing connections among findings obtained under different study conditions. Source attribution makes individual findings trace…
2026-11-02
- PixelBench: Can LLMs Understand and Generate Images Pixel by Pixel?
Large language models (LLMs) can generate visual assets through text-based formats such as SVG and CAD, but these symbolic representations describe images using high-level primitiv…
2026-11-02
- Tune the Baseline First: Most of Muon's Scaled Weight Decay Gain Is a Tuning Gap, but Its Small Real Advantage Grows With Model Size
Scaled weight decay for Muon (Muon-SW) has been reported to train language models more efficiently than constant decoupled weight decay, but such comparisons usually reuse one lear…
2026-11-02
- Ranked but Not Stopped: Detecting Prompt-Injection Hijacks from Typed Decision Models' Output Probabilities
Programs that use a language model as a classifier act on its most probable label. Text injected into the input can move that label to one the attacker chooses, a hijack. A monitor…
2026-11-02
- Same Sentence, Different Construct: Identity-Swapped Rationales Reveal Measurement Non-Invariance in LLM Hate Speech Classifiers
Large language models are increasingly used as measurement instruments for socially consequential constructs such as hate speech, and are typically validated by their aggregate agr…
2026-11-02
- Chain on Demand: Deferred Evidence Organization for Long-Term Conversational Memory
Long-term conversational memory faces a timing mismatch: write-time summaries and graphs organize history before a future query is known, while independent retrieval can allocate a…
2026-11-02
- GigaSpeech 3: Scaling Multi-Dialect Arabic Speech Recognition with Weakly Labeled Data
Dialectal Arabic speech recognition remains constrained by the scarcity of dialect-resolved transcribed speech despite the abundance of web audio. We introduce GigaSpeech 3, a corp…
2026-11-02
- As Expected? Efficient LLM Planning via Expectation-Observation Consistency Verification
LLM planning often relies on reactive executors (e.g., ReAct) that generate a single action at each step to remain grounded in the latest observations, incurring substantial infere…
2026-11-02
- Diagnosing Model--Item Interactions in Language Model Benchmarks with LSIRM
Language models with similar benchmark accuracy can succeed on different items. Overall scores do not show where their performance differs or which item properties might explain th…
2026-11-02
- InteractionBench: Measuring Timing, Task Accuracy, Whisper and Volume, and Grounding in Real-Time Voice Systems
Most benchmarks for real-time voice systems focus on one aspect, such as turn-taking, instruction following, paralinguistic control, or conversational memory, each with its own dat…
2026-11-02
- When Rewards Depend on the Group, GRPO Plays a Tournament
Rewards for diversity, coverage and distribution matching make the reward of each sampled response depend on the rest of its group, and they are now routinely optimized with GRPO a…
2026-11-02
- Solver-Like Behavior in Large Language Models as Formalizers
In the LLM-as-formalizer paradigm, a large language model translates a problem into a formal program and leaves the search to a solver, gaining the solver's guarantees. Formalizers…
2026-11-02
- EviCast: Forecasting Retrieval Gains Through Shared Response Calibration
Retrieval-augmented generation (RAG) extends large language models with external evidence, yet relevant evidence does not ensure that a generator can use it. Indiscriminate retriev…
2026-11-02
- Beyond Accuracy: How Quantization Affects Coding Agents
Agents are increasingly used for coding tasks, but the inference cost of large language models (LLMs) limits where they can run. Quantization reduces model size, but its effects on…
2026-11-02
- VisCodeBench: a new benchmark dataset for Text-to-Vis reflecting real-world practice
Generating visualisations from natural language requests (text-to-vis) enables users to explore data more efficiently. While LLMs can complete simple visualisation requests, their …
2026-11-02
- Which Words a Model Fears: From a Refusal Direction to a Vocabulary-Wide Fear Lexicon
Whether an aligned language model refuses depends in part on which words a prompt contains, regardless of whether those words carry the request's meaning. Harm words shift the hidd…
2026-11-02
- Flide: Accelerating Decision Inference with Exact State Forking
Decision models such as Jev have emerged as a highly promising paradigm for LLM agent systems, providing lightweight and reliable categorical reasoning for rapid System One executi…
2026-10-30
- DecisionRank: On-Policy Distillation for Decision-Model Rerankers
Decision models target structured tasks by directly predicting choices or scores, avoiding autoregressive decoding latency. However, transferring a generative LLM's ranking knowled…
2026-10-30
- RangeSWA: Training-Free Length Extrapolation for Continuous Diffusion Language Models
Continuous diffusion language models show promising potential in text generation tasks, but are prone to problems such as text repetition, semantic drift, and generation quality de…
2026-10-30
- PersonaChatBench: Evaluating Chat Assistants with Diverse Simulated Users
Conversational assistants are expected to help people with diverse backgrounds, preferences, and communication styles. Yet evaluations centered on a single cooperative user provide…
2026-10-30
- Asynchronous Flow Map Language Models
The order in which tokens are predicted in language modeling has a profound impact on the quality of generated sequences. Autoregressive and masked diffusion models iteratively fil…
2026-10-30
- TurnTurk-GPT: Text-Based Turn-Taking Prediction for Turkish
This paper presents TurnTurk-GPT, a text-based turn-taking predictor for Turkish obtained by fine-tuning a Turkish GPT-2 model with the TurnGPT objective on synthetic and real dial…
2026-10-30
- Less Loss, Fixed Weights: Logit Corrections for Forgetting
When a language model is fine-tuned to follow instructions, or further pretrained on a new subject, it often gets worse at predicting standard web text. This is known as "forgettin…
2026-10-30
- Aligning MLLMs with the Latent Structure of Human Cognition via Behavior-Derived Semantic Dimensions
Multimodal Large Language Models (MLLMs) have achieved impressive performance across a wide range of reasoning tasks, yet it remains unclear whether their internal representations …
2026-10-30
- Completeness and Cost of Retrieval Structures for Multi-Tool Requests in LLM Agents
An agent that must call several tools to fulfill a request fails when the tool retriever omits any one of them, yet tool retrievers are mostly built and evaluated on queries requir…
2026-10-30
- How Much Can Tool Calls Reveal About You? Privacy Leakage from Tool-Call Traces
Large language model (LLM)-based agents interact with external tools through protocols such as the Model Context Protocol (MCP), producing tool-call traces that may expose behavior…
2026-10-30
- Text as an Index for Zero-Shot Cross-Domain Knowledge Tracing
Knowledge tracing (KT) predicts a student's next response from their history. New platforms often have question texts but no response logs for training, motivating zero-shot cross-…
2026-10-30
- SimXAI: A Simple Drop-in Layer Increases Interaction Quality and Understanding in Conversational XAI
Conversational explainable AI (ConvXAI) agents often use GUS-style architectures with separate modules for intent recognition, dialogue management, and template-based response gene…
2026-10-30
- Iterated Integrals as Lexical Fingerprints: Path Signature Embeddings for Lexical Similarity
While sentence embeddings are primarily designed for semantic generalization, their latent geometry implicitly encodes surface-level lexical properties, a relationship that remains…
2026-10-30
- Peeling the Matryoshka: Extracting Ontological Structure from Retrieval Embeddings
Taxonomy construction is essential for knowledge-intensive NLP yet remains costly: clustering-based methods (i) need careful per-corpus tuning and yield unstable hierarchies, while…
2026-10-30
- Stepwise-Querying: Distilling Reasoning without Access to Teacher Chain-of-Thought
Reasoning distillation is widely used to transfer the reasoning capabilities of large language models to smaller students through teacher-generated supervision. In the black-box se…
2026-10-30
- Elucidating the Design Space of LM Data Attribution
Data attribution methods for Language Models (LMs) are typically evaluated using downstream tasks like fact retrieval. However, existing evaluations vary multiple dataset propertie…
2026-10-29
- ShallowDefend: Harnessing Shallow Safety Alignment for Multi-Turn Defender
Safety alignment has enabled large language models to reliably refuse harmful requests directly, yet the same request distributed across a multi-turn conversation still succeeds fa…
2026-10-29
- Testimony from Timelines: Evidence-Grounded Significance Testing of Temporal Clinical Hypotheses
Ask a language model whether sales tend to dip after a product recall, or whether one side effect tends to follow another after a vaccine, and it answers in a fluent sentence that …
2026-10-29
- Multi-View MoE: Letting Every Router Shape Expert Computation
Sparse mixture-of-experts (MoE) models expand parameter capacity by activating a small subset of experts per token. Expert selection directly affects model capabilities under a fix…
2026-10-29
- PIVOT: Progressive Intervention via Online Teacher Correction
Reinforcement learning from verifiable rewards (RLVR) is limited by sparse outcome-level feedback on hard reasoning tasks: when all rollouts in a group fail, rewards provide no lea…
2026-10-29
- Every Token Needs Its Own Teacher: Multi-teacher Adaptive On-policy Self-distillation
On-policy Self-distillation (OPSD) has proven effective for post-training large language models by constructing a teacher conditioned on privileged references. However, existing ap…
2026-10-29
- IndicIPR-QA: A Benchmark for Extractive Question Answering on Indian IPR Laws
Legal Question Answering (QA) over statutory texts is a challenging task in Natural Language Processing (NLP), especially in low-resource settings with complex legal systems. It re…
2026-10-29
- A Low-Rank Space Captures Linear Alignment Between LLM and Human Neural Activity During Naturalistic Language Listening
Large language models (LLMs) are increasingly used as computational models of human language processing, yet it remains unclear which components of their representations drive alig…
2026-10-29
- Agentic University: Collaborative Agents for Personalized Learning
Large language models support personalized tutoring, content generation, and adaptive learning paths. However, capability development requires these resources and interactions to f…
2026-10-29
- rTaste: Do Research Agents Have Taste? Evaluating Early Stage Research Judgment
Do research agents have taste, the judgment of which experiments are worth running before their outcomes are known? We give six frontier models seven recent NLP papers. Each model,…
2026-10-29
- AI Cosmos: An Agentic Infra for Situated Learning
Large language models (LLMs) can improve task performance during learning, yet these gains do not necessarily reflect learners' ability to thereafter apply the knowledge and skills…
2026-10-29
- Can LLMs Ignore What They Know? Measuring Knowledge Leakage in Persona-Based User Simulation
When persona-conditioned responses are read as respondent-level simulations, each response should reflect the described persona's knowledge, and the respondent written to know noth…
2026-10-29
- Prompts and probabilities in the linguistic assessment of language models: Evidence from semantic focus-sensitivity
As language models (LMs) have improved, an increasingly common means of evaluating their linguistic competence is through explicit prompting (e.g., "Which sentence is more acceptab…
2026-10-29
- Tokenizer Tokens Tell Training Tales
Tokenizers are usually treated as fixed preprocessing tools, but they are trained on text and can retain traces of that text in their vocabularies. We ask whether a BPE tokenizer c…
2026-10-29
- KV Cache Quantization for Diffusion Language Models
KV cache quantization is well studied for autoregressive language models, where cached prefix states are exact under causal attention and quantization is mainly a cache reconstruct…
2026-10-29
- Grammar Control for Language Learning with Sequential Monte Carlo and Small Language Models
Second language learners benefit from exposure to level-appropriate grammar in interactive language use. While prompted frontier language models can usually produce target grammar …
2026-10-29
- The Role of Self-Concept In Language Models
When a language model says "I," does it refer to its parameter weights, its currently running process, its stable preferences, its model lineage, or the goals it pursues? Different…
2026-10-29
- When the Top Answer Decides: Gender Bias in Decision Models
Decision models are language models built to route, classify and triage: they return option probabilities from one forward pass, and an application acts on the top answer. They are…
2026-10-29
- From the Transformer to a Formal Semantics of Coordinated Transformations
We develop a formal semantics for a general class of coordinated systems of parameterized transformations, taking the Transformer as a paradigmatic starting point but abstracting f…
2026-10-29
- Event-SAMA: Language-Grounded Video Object Segmentation on Neuromorphic Event Streams
Event cameras offer microsecond temporal resolution, high dynamic range and no motion blur, but referring video object segmentation (RVOS) lacks event-stream datasets and models. W…
2026-10-29
- Incentivizing Self-Learning in Large Language Models via Reinforcement Learning
Large language models (LLMs) draw on parametric knowledge and retrieval-augmented generation (RAG) to solve knowledge-intensive tasks. Yet problem solving in unfamiliar domains cal…
2026-10-29
- Evaluating Cultural Knowledge Editing in Vision-Language Models
Current Vision-Language Models (VLMs) frequently fail to recognize culture-specific concepts, especially from underrepresented regions. Effectively injecting this long-tail knowled…
2026-10-29
- Red-Teaming Large Language Models in Low-Resource African Languages: A Case Study of Yoruba and Nigerian Pidgin
As large language models (LLMs) become more widespread in Africa, their safety in low-resource languages remains largely untested since most red-teaming work focuses on English. Th…
2026-10-29