Search: “models”

Feature Elicitation for SAE Steering of Deceptive Behaviour in Small LLMs

arXiv:2610.20638v1 Announce Type: new Abstract: Research on deceptive alignment is becoming increasingly urgent as frontier-model capabilities advance and their risk of causing catastrophic harm incre

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

Fractional Power-of-Two Quantization for Efficient and Effective Multiplier-Free LLM Inference

DGX agent

arXiv:2610.21257v1 Announce Type: cross Abstract: Large Language Models (LLMs) demand efficient inference, where the multiply-accumulate (MAC) operations in linear projections dominate compute and ene

model-releasesarxiv-cs-ai

16 Oct 2026

Tutorials

Full-Duplex Interaction with Speech, Expressions and Gestures

arXiv:2610.21452v1 Announce Type: cross Abstract: Human face-to-face conversation requires coordinating speech, gestures and expressions across both speaking and listening. We present ELLSA 2, an end-

tutorialsarxiv-cs-ai

Research

Lexical Reference Frames: Interpreting LM Residual Streams in Native Semantics

arXiv:2610.22169v1 Announce Type: new Abstract: The vocabulary of a language model is its native semantics: the embedding and the unembedding fix the meaning of a state at the two ends of the residual

researcharxiv-cs-cl

Agents

Mallea: Building, Steering, and Memorizing Worlds with Coding Agent

arXiv:2610.21283v1 Announce Type: new Abstract: Code world models represent world dynamics through explicit state and executable rules. We investigate how to coordinate object representations, behavio

agentsarxiv-cs-ai

Measuring the local shape of learning-rate transfer curves

arXiv:2610.21814v1 Announce Type: new Abstract: Understanding the learning-rate (LR) dependence of trained model performance is essential for better optimization of neural networks, but the appropriat

Memorilla: Latent Semantic Memory for LLMs

arXiv:2610.21375v1 Announce Type: cross Abstract: Memory enables large language models (LLMs) to use information beyond their parameters, such as past user conversations, long documents, or an agent's

Memory as Weights: Internalizing Long-Term History for Streaming Videos

arXiv:2610.22070v1 Announce Type: cross Abstract: Streaming vision-language models (VLMs) must continuously integrate information from the past while remaining responsive to current observations and u

Applications

Multi-task Neural Diffusion Processes

arXiv:2510.03419v3 Announce Type: replace-cross Abstract: Neural diffusion processes provide a scalable, non-Gaussian approach to modelling distributions over functions, but existing formulations are

applicationsarxiv-cs-ai

Neither True nor False: Do LLMs Represent Truth-Value Gaps in Plural Predication?

arXiv:2610.14438v2 Announce Type: replace Abstract: Large Language Models (LLMs) are increasingly being used to summarize and verify claims in critical sources that inform consequential decisions. Suc

safetyarxiv-cs-cl

One Bias Score, Two Factors: How Mitigations Affect Option Bias in Multiple-Choice Evaluation

arXiv:2610.20993v1 Announce Type: new Abstract: Reordering multiple-choice options can change a model's answer, an effect studied as selection bias (a preference for certain option letters) and as ord

safetyarxiv-cs-cl

PremiseCheck: Efficient Fact-Checking of Human and AI Peer Reviews

arXiv:2610.21046v1 Announce Type: cross Abstract: Peer review is under strain from the growth of submissions, and review comments written by humans or AI can misstate what a paper says. We introduce R

PriorityFEC: Deadline-Aware Streaming FEC for KV-Cache Handoff in Edge Multi-Agent Systems

arXiv:2610.21390v1 Announce Type: cross Abstract: Multi-agent large language model (LLM) systems are increasingly deployed at the edge to support complex tasks that require collaboration and informati

REKI: Registering Knowledge in VLAs

arXiv:2610.22178v1 Announce Type: cross Abstract: We introduce REKI, a 2.4B vision-language-action model (VLA) pretrained exclusively on public data that sets a new Pareto frontier between performance

safetyarxiv-cs-cv

Research

Rethinking Routing Exploration in Modern Mixture-of-Experts

arXiv:2610.22353v1 Announce Type: cross Abstract: Mixture-of-experts (MoE) models have become an important approach for scaling model parameters. Modern MoE architectures increasingly combine two desi

researcharxiv-cs-ai

Score Mismatch and Set-Constrained Sampling in Zero-Shot Conditional Discrete Diffusion

arXiv:2610.21061v1 Announce Type: cross Abstract: Discrete diffusion models have become a powerful framework for modeling complex discrete distributions across a range of applications. While theoretic

safetyarxiv-cs-lg

SSR: Structural Signal Retriever for Scalable Agent Discovery in LLM-Orchestrated Multi-Agent Systems

arXiv:2610.21113v1 Announce Type: cross Abstract: Large-scale multi-agent ecosystems require mechanisms for discovering suitable agents without exposing entire registries to a large language model (LL

Applications

Task-Centric Multimodal Learning for Cognitive Impairment Detection

arXiv:2610.20906v1 Announce Type: cross Abstract: Dementia affects millions of older adults worldwide, yet early detection remains difficult outside clinical settings. Traditional models primarily rel

applicationsarxiv-cs-cl

TimeCap: A Scalable Open-Source Pipeline for Captioning Wearable Health Time-Series Data

arXiv:2610.21274v1 Announce Type: cross Abstract: Time-series data generated continuously across industrial, environmental, and especially health domains produce multimodal signals whose scale and het

Topology-Inspired Fine-Tuning for Wind-Field Super-Resolution

arXiv:2610.22262v1 Announce Type: new Abstract: Scientific super-resolution (SR) models are typically optimized with pointwise reconstruction losses, yet scientific interpretation often hinges on extr

Translation Circuit Discovery and Targeted Optimization for Low-Resource Multilingual Translation

arXiv:2610.21456v1 Announce Type: new Abstract: Recent multilingual large language models achieve strong translation performance, yet the internal mechanisms responsible for low-resource multilingual

Twice Upon a Time: Complementary Representations for Time-Series Reasoning

arXiv:2610.22327v1 Announce Type: cross Abstract: Time Series Language Models (TSLMs) enable language-based understanding and reasoning over temporal data, yet a fundamental question remains: what inf

safetyarxiv-cs-ai

Research

Two Sample Sizes in Test-Time LLM Ensembling: Buffered ERM and Minimax Theory

arXiv:2610.21363v1 Announce Type: cross Abstract: Test-time scaling can improve the performance of large language models by generating multiple answers to a question and aggregating them. Weighted ens

researcharxiv-cs-lg

VisualTrust: Video Reasoning under On-Screen Claim Framing

model-releasesarxiv-cs-cv

arXiv:2610.21159v1 Announce Type: new Abstract: On-screen text can describe what a video shows, or it can frame what the video is supposed to mean. Current benchmarks treat both alike, testing whether

Who May Release an Agent's Promise? Standing over Commitments in Multi-Party Chat

arXiv:2610.21203v1 Announce Type: new Abstract: When an assistant serving a team makes a promise to Alice, who may cancel it? We call this standing: only the creditor (the person owed) or someone the

A Decade in a Day: Synthesizing Agent-Native Interfaces from Human-Oriented Software Evolution

arXiv:2610.15765v1 Announce Type: cross Abstract: Large language model agents are emerging as an important new class of users of desktop applications. However, existing desktop interfaces were designe

Adaptive Observation Control for Active Visual Recognition

model-releasesarxiv-cs-cv

arXiv:2610.19624v1 Announce Type: new Abstract: Visual recognition models usually process a fixed view under a fixed computation budget, although images often benefit from different scales and amounts

AFCS: Atomic-Fact-Conditioned Synthesis for Factual Knowledge Pretraining

arXiv:2610.20086v1 Announce Type: new Abstract: Synthetic data is increasingly used in language-model pretraining. Existing knowledge-corpus synthesis pipelines, including WRAP and Nemotron-CC, usuall

AssetForge: Asset-Augmented Task Synthesis for Business Automation Agents

arXiv:2610.18330v1 Announce Type: new Abstract: Training business automation agents requires executable tasks that connect a natural-language request to an application state and a verifiable end state

Bayesian Causal Forests for Zero-Inflated Continuous Outcomes

arXiv:2610.18889v1 Announce Type: cross Abstract: Estimating (heterogeneous) treatment effects is still challenging when the outcome is semicontinuous (i.e., exhibiting both a discrete point mass at z

Behavioral History Outperforms Descriptions of the Person for LLM Synthetic Personas

arXiv:2610.03998v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used as synthetic personas representing survey respondents. Their validity as substitutes for particul

BenchPreS: A Benchmark for Context-Aware Personalized Preference Selectivity of Persistent-Memory LLMs

arXiv:2603.16557v2 Announce Type: replace-cross Abstract: Large language models (LLMs) increasingly store user preferences in persistent memory to support personalization across interactions. However,

Beyond Static Cues: Unified Dynamic Detection of MLLM Backdoors via Dual-Flow Trajectory Analysis

arXiv:2610.20319v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) are increasingly deployed in safety-critical domains, yet their susceptibility to backdoor attacks remains a fo

safetyarxiv-cs-cv

Can LLM Agents Efficiently Identify Appliance Faults? Insights From Diagnosing a Simulated Appliance

arXiv:2610.19291v1 Announce Type: new Abstract: In so-called 'Industry 4.0,' integrated sensors, analytics tools, and other digital technologies supplement industrial workflows, potentiating enhanced

Correct Choices, Failed Sequences: Choice2Seq Benchmarks Correctness Retention Across Output Interfaces

arXiv:2610.16191v1 Announce Type: new Abstract: A model may select the correct answer yet fail to construct it through a sequence of local decisions. We introduce Choice2Seq, a diagnostic benchmark th

Cowork-Gym: Synthesizing Interactive Work Environments at Scale

arXiv:2610.17670v1 Announce Type: new Abstract: Cowork agents navigate dependencies across tools and files while adapting to changing user requirements and edits to shared workspaces. Scaling such env

Diagnosing and Correcting Reward Hacking in Verifiable Instruction Following

arXiv:2610.19525v1 Announce Type: cross Abstract: Verifiable instruction-following benchmarks emphasize output-constraint satisfaction but can overlook whether responses meaningfully answer the underl

Agents

Do Perturbation Predictors Learn Gene Interactions? Diagnosing Extrapolation in Agentic Screening

arXiv:2610.18880v1 Announce Type: new Abstract: High-throughput perturbation screens increasingly use machine-learning models to predict the effects of gene combinations that were never measured. A di

agentsarxiv-cs-lg

Double-Rollout Reinforcement Learning with Sampling-based Reward Estimation

arXiv:2610.16593v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) implemented with Group Relative Policy Optimization (GRPO) has driven substantial progress in trai

Exhaustive Canonical Skeletons Generation: A Formal Dataset for Decomposable Symbolic Regression

arXiv:2610.20108v1 Announce Type: new Abstract: Neural symbolic regression (SR) models require diverse datasets of mathematical expressions for robust training. However, existing datasets rely predomi

Local Ai

From Plato to Aristotle: Toward a General Theory of Representation Hypotheses

arXiv:2610.20286v1 Announce Type: new Abstract: The Platonic Representation Hypothesis (PRH) holds that increasingly capable models converge toward a common representation. However, recent studies hav

local-aiarxiv-cs-lg

From Triage to Impact: A Multi-Hop Reasoning Benchmark for Explainable Persian Hate Speech

arXiv:2610.20370v1 Announce Type: new Abstract: Hate speech has become increasingly prevalent on social media, raising concerns about its detection, interpretation, and moderation. Persian, a relative

Harness Policy Distillation: Aligning Agent Policies with Runtime Semantics

arXiv:2610.20261v1 Announce Type: new Abstract: Deploying a general-purpose model as an agent means running it inside a product-specific runtime harness that decides which actions are executable, how

safetyarxiv-cs-lg

Agents

L3M: A Verifiable Harness and Runtime for Self-Extending and Self-Improving Agents

arXiv:2610.19591v1 Announce Type: cross Abstract: The lasting gains in AI have come from methods that absorb more computation, and language models already scale computation within a call. Useful work,

agentsarxiv-cs-lg

Tutorials

Learning How to Remember: Memory Modality Selection for GUI Agents

arXiv:2610.17928v1 Announce Type: new Abstract: Recent advances in vision-language models (VLMs) have substantially improved the capabilities of GUI agents, while memory augmentation has emerged as a

tutorialsarxiv-cs-ai

Learning to Compute Is Not Learning to Retain or Reuse in Looped Transformers

arXiv:2610.19768v1 Announce Type: new Abstract: Looped transformers compute by applying one block repeatedly, so they can run for more loops than they were trained with. Recurrent reasoners are known

LIBERO-Industry: Benchmarking Robotic Manipulation under Industrial Process Constraints

arXiv:2610.16782v1 Announce Type: new Abstract: Industrial manipulation in flexible manufacturing, e.g., high-mix, low-volume (HMLV) electronics assembly, requires robot policies to adapt to changing

Loose Lips: Understanding Privacy Leakage in Agentic Dialogue

arXiv:2610.14071v1 Announce Type: cross Abstract: As LLM agents are increasingly deployed in high-stakes enterprise settings, understanding how they handle sensitive information in agentic dialogue ha