5
Interaction Protocol Shapes Moral Judgment in Multi-Agent Debate
A study demonstrates that the interaction protocols used in multi-agent debates significantly influence the moral reasoning and judgment outcomes of large language models.
5
Experience Memory Graph: One-Shot Error Correction for Agents
A new error-correction method called Experience Memory Graph aims to help LLM agents recover from failures in long-horizon tasks more efficiently than traditional reflection techniques.
5
Interventional Grounding Audits: Black-Box Premise-Dependency Tests for LLM Chain-of-Thought via Predicate Substitution
A new auditing method uses predicate substitution to test whether large language models genuinely rely on stated premises during chain-of-thought reasoning.
5
DIVE: Embedding Compression via Self-Limiting Gradient Updates
DIVE is a new dimensionality reduction technique for language model embeddings that uses self-limiting gradient updates to improve compression efficiency.
5
From Language to Navigation Goals: A Vision-Language Approach for Semantic Navigation of Mobile Robots Using RGB-D Perception
A new framework enables mobile robots to interpret natural language commands for autonomous navigation using RGB-D perception.
5
MedDiffuseMix: Preserving Diagnostic Evidence with Saliency-Aware Diffusion Medical Image Data Augmentation
MedDiffuseMix provides a saliency-guided diffusion framework to augment medical imaging data while preserving critical diagnostic features.
5
Learning to Learn-at-Test-Time: Language Agents with Learnable Adaptation Policies
A new method for test-time learning introduces learnable adaptation policies to help language agents improve their performance through iterative interaction.
5
When Audio Separation Hurts Zero-Shot ASR: Evaluating SAM-Audio with Whisper on Bengali and English Speech
The study evaluates the impact of audio separation preprocessing on zero-shot automatic speech recognition performance, finding that cleaner audio does not always improve transcription accuracy.
5
MASPRM: Multi-Agent System Process Reward Model
The Multi-Agent System Process Reward Model provides a method to evaluate and optimize message sequences between agents during inference-time search.
5
A Causal Model of Theory of Mind in Conflict for Artificial Intelligence
This paper proposes a causal framework to determine when artificial intelligence systems should engage in theory of mind processes during conflict scenarios.
5
Generative Compilation: On-the-Fly Compiler Feedback as AI Generates Code
The authors present a method for integrating compiler feedback directly into the autoregressive decoding process to improve the quality of AI-generated code.
5
Unleashing Multimodal Large Language Models for Training-free HOI Detection in the Wild
Researchers have proposed a training-free approach for detecting human-object interactions in the wild by leveraging the capabilities of multimodal large language models.
5
Traffic-Aware Randomized Smoothing for LLM-Based Network Intrusion Detection
A new defense mechanism called Traffic-Aware Randomized Smoothing has been proposed to improve the robustness of LLM-based network intrusion detection systems against traffic manipulation.
5
Anatomically Faithful but Temporally Blind: Auditing Attribution for Left-Ventricular Ejection-Fraction Estimation from Echocardiography
A study evaluating deep learning models for echocardiography analysis reveals that current attribution methods often fail to verify temporal faithfulness in medical imaging predictions.
5
GHR-VLM: Making Zero-Shot Transit Video Analytics Realizable with Grounded Hybrid Reasoning
The GHR-VLM framework combines grounded reasoning with vision-language models to enable zero-shot video analytics for transit systems without requiring task-specific training data.
5
ExTernD: Expanded-Rank Ternary Decomposition Ternary LLM PTQ with Accuracy Approaching Any Quantization Level
The ExTernD method introduces a post-training factorization technique for large language models that utilizes expanded-rank ternary decomposition to improve quantization efficiency and accuracy.
5
Discrete Diffusion Models: A Unified Framework from Tokenization to Generation
A new framework for discrete diffusion models explores how tokenization and vocabulary structure influence generative performance.
5
ScanFocus: A Coarse-to-Fine Framework for Spatio-Temporal Video Grounding
The proposed ScanFocus framework addresses computational efficiency and precision in spatio-temporal video grounding through a coarse-to-fine processing approach.
5
Learning Engagement Assistant (LEA): Cross-Course Scalability and Classroom Evaluation of an Agentic AI Tutoring System
A study evaluates the scalability and classroom performance of an AI tutoring agent that integrates retrieval-augmented generation with structured knowledge models.
5
Faithful Autoformalization of Natural Language Assertions
The authors introduced Monty, a framework designed to improve the accuracy of autoformalization by converting natural language specifications into executable software assertions.
5
Reassessing Muon for Matrix Factorization
This paper provides a theoretical analysis of the Muon optimizer to clarify the mechanisms behind its performance advantages in large-scale deep learning.
5
SteinGate: Tail-Sensitive Safe Reinforcement Learning via Stein Discrepancy
The SteinGate framework introduces a new safety certification method for reinforcement learning to better mitigate rare but catastrophic risks.
5
STKAN: Kolmogorov-Arnold Networks for Spatio-Temporal Forecasting
The authors introduce STKAN, a spatio-temporal forecasting architecture that leverages Kolmogorov-Arnold Networks to better model complex traffic data.
5
The Hitchhiker's Guide to Monoculture
An analysis of Kaggle contest submissions examines whether the use of AI coding assistants is leading to increased homogenization of software development outputs.