All papers

The paper proposes an in-context reinforcement learning method enabling vision-language-action policies to adapt during robot deployment.

This paper introduces a multi-hop reasoning benchmark to evaluate fine-grained, explainable hate speech analysis in Persian.

The paper introduces a distillation framework that triggers teacher takeovers when student policies diverge during long-horizon reasoning.

The paper proposes a hierarchical two-loop optimization framework that uses residual errors to trigger active evidence acquisition during inference.

The paper presents a multimodal agent designed to trace the origin, spread, and potential safety risks of online memes.

The paper presents an online Thompson Sampling bandit policy to select classifiers that balance predictive accuracy and subgroup fairness under delayed feedback.

Presents an end-to-end distributed system orchestrating coding agents to execute computational research experiments alongside human scientists.

This paper establishes theoretical limits on what forms of annotator disagreement can be identified from pairwise preference feedback data.

The paper presents a graph-based framework using control nodes and behavior trees to coordinate heterogeneous artificial intelligence models.

Proposes conformal calibration to weight criterion-level uncertainty when using rubric-based language model judges for reinforcement learning.

The paper introduces a hierarchical task network framework that separates multi-agent planning from execution scheduling.

The paper introduces a parameter decomposition method to interpret the mechanistic subcomponents of language model weights.

The paper uses language model subtask decomposition and dense progress tracking to enhance reinforcement learning on long-horizon tasks.

The paper examines AI-generated text detection in scientific papers by combining visual formatting, PDF text, and LaTeX representations.

The authors present an agent loop that closes functional test coverage in hardware design using independent simulation scoreboards.

The paper derives mathematical bounds to evaluate the calibration of query-based probabilistic models using logical consistency without ground-truth labels.

Develops a recursive self-improvement framework where an evolving meta-agent adapts agent workflows based on execution trajectories and feedback.

The paper investigates data poisoning attacks targeting pre-training and retrieval corpora of protein language models and generative design systems.

Develops a domain harness and policy learning pipeline to improve language model agent execution in computer-aided design environments.

Improves deep search agents by refining search queries and filtering evidence to align retrieval with multi-step reasoning goals.

Presents a bound-propagation verification framework to bound worst-case reconstruction errors for deep joint source-channel coding decoders under noise.

Combines language model dialogue with workflow automation to process satellite imagery for internal solitary wave detection and forecasting.

This paper investigates internal model representations that separate assistant actions from externally provided text across conversational layers.

Uses tool-augmented language models to extract clinical variables from patient notes and reliably execute deterministic medical calculation formulas.

Introduces a Ukrainian meme dataset and evaluates vision-language models on detecting persuasive and propaganda techniques.

Analyzes how prompt variations redistribute internal feature activations in sparse autoencoders and traces their impact on downstream model decisions.

Proposes a training framework enabling aerial vision-language agents to navigate in full action chunks and stop reliably at destinations.

Integrates reinforcement learning and tabular model guidance to help language models select features and predict emergency department clinical risk.

The paper evaluates the impact of sarcasm on Turkish hate speech detection models using a targeted external test dataset.

The paper presents an agent system that generates, edits, and renders interactive virtual worlds from text using executable code rules.

Develops a closed-loop distillation pipeline where student feedback refines teacher models to improve adversarial robustness.

The paper presents a metric and benchmark measuring how quickly environment dynamics can change before agent decision-making fails.

Proposes a training-free structural lexical retriever to select relevant agent cards without overloading model context windows.

Uses reinforcement learning to train an agent module that selects, sizes, and orders retrieved memories for downstream tasks.

The paper proposes an algorithm using betting-based sequential tests to monitor supermartingale certificate violations during runtime execution of stochastic systems.

Presents a multi-agent system for paper screening and data extraction in systematic literature reviews following standard guidelines.

The paper presents a neural-symbolic method for multi-robot systems to evaluate robot and object hypothesis trust purely via inter-robot observation consistency.

Introduces a multimodal harness and benchmark pairing continuous video with clinical records to assist patient monitoring in hospitals.

Introduces an imprecise Dirichlet model method to calculate interval-valued Shapley values and analyze attribution robustness in tree-based models.

The paper proposes an adaptive skill framework that extracts patterns from trajectories to enable multi-modal agents to self-evolve.

Develops a personalized federated preference optimization framework maintaining local LoRA adapters atop a globally synchronized model.

The paper introduces an agentic framework combining language model reasoning with symbolic engines for formal inequality proving in Lean4.

Develops robust estimators and optimization algorithms to learn linear dynamical systems from trajectories corrupted by adversarial outliers.

This paper presents a benchmark to evaluate how adversarial attacks affect reasoning security in multi-agent debate systems.

The paper presents an evidence-driven multi-agent framework for collaborative vulnerability understanding and automated repository-level patch generation.

The paper analyzes 33,000 GitHub pull requests to examine why AI coding agent contributions succeed or fail.

The paper introduces a logit adjustment method based on pairwise comparisons between sampled reasoning paths to improve test-time answer selection in language models.

Proposes Agentless, an agent-free approach to software development issues that compares favorably against autonomous software engineering agents.

Presents an open-source library that unifies training, evaluation, reinforcement learning, and distillation workflows for tool-using language model agents.

Introduces a low-latency guardrail model that routes incoming agent actions through editable policy trees for safety auditing.

The paper introduces a system layer that separates control flow from untrusted data to protect language model agents against prompt injection attacks.

This paper introduces specification templates to define operational boundaries, authority, and oversight for autonomous software agents.

Proposes dynamic activation steering using preceding denoising predictions to reduce hallucinations in diffusion language models.

The paper presents a multi-stage automated labeling framework using multimodal models and agents to annotate short videos for content moderation.

The paper provides a conceptual taxonomy and review distinguishing task-specific AI agents from collaborative agentic AI systems.

Maps Erlang and OTP fault-tolerance principles onto multi-agent LLM systems to evaluate supervisory reliability architectures.

Investigates how changing tasks redistributes attention head influence and alters internal activations in open-weight large language models.

Introduces AgentChaosBench, a benchmark for detecting and localizing injected runtime boundary faults in LLM-based multi-agent applications.

Decomposes joint temporal logic constraint violations to allocate credit among agents in cooperative reinforcement learning.

The paper shares eight lessons from building an agent pipeline that ports legacy C/C++ code to memory-safe CHERI hardware architectures.