What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspective
Paper
• 2410.23743 • Published
• 64
Psychology
-
-
Large Language Models Orchestrating Structured Reasoning Achieve Kaggle Grandmaster Level
Paper • 2411.03562 • Published • 69 -
Polynomial Composition Activations: Unleashing the Dynamics of Large Language Models
Paper • 2411.03884 • Published • 28 -
MM-IQ: Benchmarking Human-Like Abstraction and Reasoning in Multimodal Models
Paper • 2502.00698 • Published • 24 -
Understanding R1-Zero-Like Training: A Critical Perspective
Paper • 2503.20783 • Published • 59 -
PaperBench: Evaluating AI's Ability to Replicate AI Research
Paper • 2504.01848 • Published • 37 -
VCR-Bench: A Comprehensive Evaluation Framework for Video Chain-of-Thought Reasoning
Paper • 2504.07956 • Published • 46 -
Perception Encoder: The best visual embeddings are not at the output of the network
Paper • 2504.13181 • Published • 35 -
Set You Straight: Auto-Steering Denoising Trajectories to Sidestep Unwanted Concepts
Paper • 2504.12782 • Published • 3 -
Multi-Token Attention
Paper • 2504.00927 • Published • 56 -
PATS: Process-Level Adaptive Thinking Mode Switching
Paper • 2505.19250 • Published • 46 -
Embodied Agents Meet Personalization: Exploring Memory Utilization for Personalized Assistance
Paper • 2505.16348 • Published • 52 -
Flex-Judge: Think Once, Judge Anywhere
Paper • 2505.18601 • Published • 27 -
CoMemo: LVLMs Need Image Context with Image Memory
Paper • 2506.06279 • Published • 8 -
Steering Conceptual Bias via Transformer Latent-Subspace Activation
Paper • 2506.18887 • Published • 6 -
CommVQ: Commutative Vector Quantization for KV Cache Compression
Paper • 2506.18879 • Published • 5 -
Vision-Guided Chunking Is All You Need: Enhancing RAG with Multimodal Document Understanding
Paper • 2506.16035 • Published • 89 -
Outlier-Safe Pre-Training for Robust 4-Bit Quantization of Large Language Models
Paper • 2506.19697 • Published • 44 -
OctoThinker: Mid-training Incentivizes Reinforcement Learning Scaling
Paper • 2506.20512 • Published • 47 -
Where to find Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test
Paper • 2506.21551 • Published • 28 -
MoCa: Modality-aware Continual Pre-training Makes Better Bidirectional Multimodal Embeddings
Paper • 2506.23115 • Published • 36 -
Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning
Paper • 2507.00432 • Published • 79 -
ExGRPO: Learning to Reason from Experience
Paper • 2510.02245 • Published • 81 -
Feature-Level Insights into Artificial Text Detection with Sparse Autoencoders
Paper • 2503.03601 • Published • 232 -
Perception, Reason, Think, and Plan: A Survey on Large Multimodal Reasoning Models
Paper • 2505.04921 • Published • 186 -
The Landscape of Agentic Reinforcement Learning for LLMs: A Survey
Paper • 2509.02547 • Published • 233 -
The Dragon Hatchling: The Missing Link between the Transformer and Models of the Brain
Paper • 2509.26507 • Published • 547 -
MMGR: Multi-Modal Generative Reasoning
Paper • 2512.14691 • Published • 119