Papers

New on arXiv

Sun 15 Feb 2026

177 papers

Context Over Compute Human-in-the-Loop Outperforms Iterative Chain-of-Thought Prompting in Interview Answer QualityLLM reasoning → Prompting and chain-of-thought reasoning
FedEMA-Distill: Exponential Moving Average Guided Knowledge Distillation for Robust Federated LearningKnowledge distillation → Federated, robust and private distillation
Trustworthy AI Posture (TAIP): A Framework for Continuous AI Assurance of Agentic Systems at Horizontal and Vertical scaleLLM agents and tool use → Agent threats, guardrails and governance
Mobile-Agent-v3.5: Multi-platform Fundamental GUI AgentsLLM agents and tool use → GUI and mobile agents
WIMLE: Uncertainty-Aware World Models with IMLE for Sample-Efficient Continuous ControlWorld models → Latent dynamics models and model-based RL +1
Zero-Shot Instruction Following in RL via Structured LTL RepresentationsDeep reinforcement learning methods → Interpretable and logic-guided RL
Data-Driven Network LQG Mean Field Games with Heterogeneous Populations via Integral Reinforcement LearningDeep reinforcement learning methods → RL for games and combinatorial optimization
Train Less, Learn More: Adaptive Efficient Rollout Optimization for Group-Based Reinforcement LearningLLM reasoning → Reinforcement learning for reasoning +1
LongCLI-Bench: A Preliminary Benchmark and Study for Long-horizon Agentic Programming in Command-Line InterfacesLLM agents and tool use → Software engineering coding agents
Conformal Signal Temporal Logic for Robust Reinforcement Learning Control: A Case StudyDeep reinforcement learning methods → Safe and constrained RL
DeepFusion: Accelerating MoE Training via Federated Knowledge Distillation from Heterogeneous Edge DevicesMixture-of-experts → Federated mixture-of-experts +1
Does Socialization Emerge in AI Agent Society? A Case Study of MoltbookLLM agents and tool use → Social simulation and multi-agent debate
AutoWebWorld: Synthesizing Infinite Verifiable Web Environments via Finite State MachinesLLM agents and tool use → Web, search, and research agents
Machine Learning as a Tool (MLAT): A Framework for Integrating Statistical ML Models as Callable Tools within LLM Agent WorkflowsLLM agents and tool use → Web, search, and research agents
KernelBlaster: Continual Cross-Task CUDA Optimization via Memory-Augmented In-Context Reinforcement LearningLLM agents and tool use → Software engineering coding agents +1
Online LLM watermark detection via e-processesLLM safety and security → Watermarking generated content and models
MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol AgentsLLM agents and tool use → Agent threats, guardrails and governance
STATe-of-Thoughts: Structured Action Templates for Tree-of-ThoughtsLLM reasoning → Test-time compute and efficient thinking
AD-Bench: A Real-World, Trajectory-Aware Advertising Analytics Benchmark for LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
GRAIL: Goal Recognition Alignment through Imitation LearningOffline RL and imitation learning from data → Imitation learning methods
Multi-Agent Debate: A Unified Agentic Framework for Tabular Anomaly DetectionLLM agents and tool use → Software engineering coding agents
Dual-Signal Adaptive KV-Cache Optimization for Long-Form Video Understanding in Vision-Language ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search AgentsLLM agents and tool use → Web, search, and research agents
Text Before Vision: Staged Knowledge Injection Matters for Agentic RLVR in Ultra-High-Resolution Remote Sensing UnderstandingLLM reasoning → Spatial and embodied reasoning +1
The Interspeech 2026 Audio Reasoning Challenge: Evaluating Reasoning Process Quality for Audio Reasoning Models and AgentsLLM reasoning → Audio-language model reasoning +1
Reasoning Language Models for complex assessments tasks: Evaluating parental cooperation from child protection case reportsLLM reasoning → Clinical and biomedical applications
SkillJect: Effectively Automating Skill-Based Prompt Injection for Skill-Enabled AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
GeoEyes: On-Demand Visual Focusing for Evidence-Grounded Understanding of Ultra-High-Resolution Remote Sensing ImageryMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
TS-Haystack: A Multi-Task Retrieval Benchmark for Long-Context Time-Series ReasoningLLM reasoning → Time series reasoning
Knowing When Not to Answer: Abstention-Aware Scientific ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning
UniWeTok: An Unified Binary Tokenizer with Codebook Size mathit2128 for Unified Multimodal Large Language ModelMultimodal LLMs / vision-language models → Visual token pruning and efficiency
Deep Dense Exploration for LLM Reinforcement Learning via Pivot-Driven ResamplingLLM reasoning → Reinforcement learning for reasoning
Index Light, Reason Deep: Deferred Visual Ingestion for Visual-Dense Document Question AnsweringLLM reasoning → Multimodal reasoning +1
When Benchmarks Lie: Evaluating Malicious Prompt Classifiers Under True Distribution ShiftLLM agents and tool use → Agent threats, guardrails and governance +2
Process-Supervised Multi-Agent Reinforcement Learning for Reliable Clinical ReasoningLLM reasoning → Clinical and biomedical applications +1
Synergistic Intra- and Cross-Layer Regularization Losses for MoE Expert SpecializationMixture-of-experts → Expert routing strategies
A Multi-Agent Framework for Medical AI: Leveraging Fine-Tuned GPT, LLaMA, and DeepSeek R1 for Evidence-Based and Bias-Aware Clinical Query ProcessingLLM reasoning → Clinical and biomedical applications +1
LaViDa-R1: Advancing Reasoning for Unified Multimodal Diffusion Language ModelsLLM reasoning → Multimodal reasoning +1
Watch the Model Think: On-Policy Extraction of Activation Steering VectorsMechanistic interpretability → Activation steering methods
DenseMLLM: Standard Multimodal LLMs for Dense PredictionMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
EgoSound: Benchmarking Sound Understanding in Egocentric VideosMultimodal LLMs / vision-language models → Video understanding with MLLMs
Toward Autonomous O-RAN: A Multi-Scale Agentic AI Framework for Real-Time Network Control and ManagementLLM agents and tool use → Agentic AI for wireless networks
Sanity Checks for Sparse Autoencoders: Do SAEs Beat Random Baselines?Mechanistic interpretability → Sparse autoencoder training and evaluation
ForgeryVCR: Visual-Centric Reasoning via Efficient Forensic Tools in MLLMs for Image Forgery Detection and LocalizationLLM reasoning → Multimodal reasoning +1
NEST: Nascent Encoded Steganographic ThoughtsLLM reasoning → Prompting and chain-of-thought reasoning +1
GUI-GENESIS: Automated Synthesis of Efficient Environments with Verifiable Rewards for GUI Agent Post-TrainingLLM agents and tool use → GUI and mobile agents
TabTracer: Monte Carlo Tree Search for Complex Table Reasoning with Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
Policy Gradient with Adaptive Entropy Annealing for Continual Fine-TuningDeep reinforcement learning methods → RL theory and sample complexity
GTS: Inference-Time Scaling of Latent Reasoning with a Learnable Gaussian Thought SamplerLLM reasoning → Test-time compute and efficient thinking
Annotation-Efficient Vision-Language Model Adaptation to the Polish Language Using the LLaVA FrameworkMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Open Rubric System: Scaling Reinforcement Learning with Pairwise Adaptive RubricLLM alignment / RLHF / preference optimisation → Reward hacking and misspecification +2
CoCoEdit: Content-Consistent Image Editing via Region Regularized Reinforcement LearningDeep reinforcement learning methods → RL with diffusion and flow models
LM-Lexicon: Improving Definition Modeling via Harmonizing Semantic ExpertsMixture-of-experts → MoE large language models
LogitsCoder: Towards Efficient Chain-of-Thought Path Search via Logits Preference Decoding for Code GenerationLLM reasoning → Test-time compute and efficient thinking
ProAct: Harnessing Streaming Motion Generation and Agentic Reasoning for Real-Time Embodied Social InteractionLLM agents and tool use → Multimodal and embodied agents
Context Shapes LLMs Retrieval-Augmented Fact-Checking EffectivenessLLM reasoning → Reasoning benchmarks and evaluation +1
Geometry-Preserving Aggregation for Mixture-of-Experts Embedding ModelsMixture-of-experts → Expert routing strategies
Choosing How to Remember: Adaptive Memory Structures for LLM AgentsLLM agents and tool use → Agent memory systems
FloCA: Towards Faithful and Logically Consistent Flowchart ReasoningLLM reasoning → Reasoning benchmarks and evaluation
BRAIN: Bayesian Reasoning via Active Inference for Agentic and Embodied Intelligence in Mobile NetworksLLM agents and tool use → Agentic AI for wireless networks
MC2Mark: Distortion-Free Multi-Bit Watermarking for Long MessagesLLM safety and security → Watermarking generated content and models
Why Self-Training Helps and Hurts: Denoising vs. Signal ForgettingKnowledge distillation → General knowledge distillation methods
GRRM: Group Relative Reward Modeling for Machine TranslationLLM reasoning → Reinforcement learning for reasoning +1
Train Short, Inference Long: Training-free Horizon Extension for Autoregressive Video GenerationDiffusion models → Video generation
From SFT to RL: Demystifying the Post-Training Pipeline for LLM-based Vulnerability DetectionLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
The Sufficiency-Conciseness Trade-off in LLM Self-Explanation from an Information Bottleneck PerspectiveLLM reasoning → Test-time compute and efficient thinking
Elastic Diffusion TransformerDiffusion models → Diffusion foundations and sampling
ATTest: Agent-Driven Tensor Testing for Deep Learning Library ModulesLLM agents and tool use → Software engineering coding agents
Chain-of-Thought Reasoning with Large Language Models for Clinical Alzheimer's Disease Assessment and DiagnosisLLM reasoning → Clinical and biomedical applications
WoVR: World Models as Reliable Simulators for Post-Training VLA Policies with RLWorld models → Embodied and robot world models +2
CodeGlance: Understanding Code Reasoning Challenges in LLMs through Multi-Dimensional Feature AnalysisLLM reasoning → Reasoning for agents, planning and code
MarsRetrieval: Benchmarking Vision-Language Models for Planetary-Scale Geospatial Retrieval on MarsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Eureka-Audio: Triggering Audio Intelligence in Compact Language ModelsMixture-of-experts → Multimodal MoE models
QuRL: Efficient Reinforcement Learning with Quantized RolloutLLM reasoning → Reinforcement learning for reasoning +1
Experiential Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
You Can Learn Tokenization End-to-End with Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
Statistical Early Stopping for Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
Why Code, Why Now: An Information-Theoretic Perspective on the Limits of Machine LearningDeep reinforcement learning methods → RL training of language and agent models
Robust Test-time Video-Text Retrieval: Benchmarking and Adapting for Query Shifts
AEG: A Baremetal Framework for AI Acceleration via Direct Hardware Access in Heterogeneous Accelerators
Full State-Space Visualisation of the 8-Puzzle: Feasibility, Design, and Educational Use
Operational Noncommutativity in Sequential Metacognitive Judgments
Automated evaluation of LLMs for effective machine translation of Mandarin Chinese to English
Empathy Is Not What Changed: Clinical Assessment of Psychological Safety Across GPT Model Generations
There Are No Silly Questions: Evaluation of Offline LLM Capabilities from a Turkish Perspective
Ethical and Explainable AI in Reusable MLOps Pipelines
FM-RME: Foundation Model Empowered Radio Map Estimation
PIPE-RDF: An LLM-Assisted Pipeline for Enterprise RDF Benchmarking
A Patient-Specific Digital Twin for Adaptive Radiotherapy of Non-Small Cell Lung Cancer
Distillation and Interpretability of Ensemble Forecasts of ENSO Phase using Entropic Learning
A Comprehensive Survey on Deep Learning-Based LiDAR Super-Resolution for Autonomous Driving
GRAFNet: Multiscale Retinal Processing via Guided Cortical Attention Feedback for Enhancing Medical Image Polyp Segmentation
An effective Genetic Programming Hyper-Heuristic for Uncertain Agile Satellite Scheduling
AXE: Grey-Box Exploitability Confirmation for Localized Vulnerability Reports
High-accuracy log-concave sampling with stochastic queries
Offline Learning of Nash Stable Coalition Structures with Possibly Overlapping Coalitions
In Transformer We Trust? A Perspective on Transformer Architecture Failure Modes
Benchmarking at the Edge of Comprehension
Floe: Federated Specialization for Real-Time LLM-SLM Inference
Differential pose optimization in descriptor space -- Combining Geometric and Photometric Methods for Motion Estimation

All 177