Top Conference Papers

Debating with More Persuasive LLMs Leads to More Truthful Answers

THE POLAR EXPRESS: OPTIMAL MATRIX SIGN METHODS AND THEIR APPLICATION TO THE MUON ALGORITHM

TRANSFORMERS ARE INHERENTLY SUCCINCT

LLMS GET LOST IN MULTI-TURN CONVERSATION

FASTER CASCADES VIA SPECULATIVE DECODING

DATA SHAPLEY IN ONE TRAINING RUN

SAFETY ALIGNMENT SHOULD BE MADE MORE THAN JUST A FEW TOKENS DEEP

LEARNING DYNAMICS OF LLM FINETUNING

TOWARDS A STATISTICAL THEORY OF DATA SELECTION UNDER WEAK SUPERVISION

THE MECHANISTIC BASIS OF DATA DEPENDENCE AND ABRUPT LEARNING IN AN IN-CONTEXT CLASSIFICATION TASK

ROBUST AGENTS LEARN CAUSAL WORLD MODELS

PROVING TEST SET CONTAMINATION IN BLACK BOX LANGUAGE MODELS

MODEL TELLS YOU WHAT TO DISCARD: ADAPTIVE KV CACHE COMPRESSION FOR LLMs

Meta Continual Learning Revisited: Implicitly Enhancing Online Hessian Approximation via Variance Reduction

IS IMAGENET WORTH 1 VIDEO? LEARNING STRONG IMAGE ENCODERS FROM 1 LONG UNLABELLED VIDEO

FLOW MATCHING ON GENERAL GEOMETRIES

BEYOND WEISFEILER-LEHMAN: A QUANTITATIVE FRAMEWORK FOR GNN EXPRESSIVENESS

APPROXIMATING NASH EQUILIBRIA IN NORMAL-FORM GAMES VIA STOCHASTIC OPTIMIZATION

AMORTIZING INTRACTABLE INFERENCE IN LARGE LANGUAGE MODELS

VISION TRANSFORMERS NEED REGISTERS

PROTEIN DISCOVERY WITH DISCRETE WALK-JUMP SAMPLING

Never Train from Scratch: FAIR COMPARISON OF LONG-SEQUENCE MODELS REQUIRES DATA-DRIVEN PRIORS

LEARNING INTERACTIVE REAL-WORLD SIMULATORS

GENERALIZATION IN DIFFUSION MODELS ARISES FROM GEOMETRY-ADAPTIVE HARMONIC REPRESENTATIONS

Spatially-Varying Autofocus

Generating Physically Stable and Buildable Brick Structures from Text

FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models

Mind the Value-Action Gap: Do LLMs Act in Alignment with Their Values?

MiCRo: Mixture Modeling and Context-aware Routing for Personalized Preference Learning