Enhong Chen

Is Softmax Loss all you need? A Principled Analysis of Softmax Loss and its Variants

Perception-R1: Advancing Multimodal Reasoning Capabilities of MLLMs via Visual Perception Reward

GTM: A General Time-series Model for Enhanced Representation Learning of Time-Series data

Evolving Graph Structured Programs for Circuit Generation with Large Language Models

From Single to Multi-Granularity: Toward Long-Term Memory Association and Selection of Conversational Agents

ChemEval: A Multi-level and Fine-grained Chemical Capability Evaluation for Large Language Models

Preference Optimization via Key-step Error Exploration for Multi-step Reasoning in LLMs

Not All Code Helps: Disentangling the Impact of Code Data on Mathematical Reasoning in Large Language Models

CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering

Benchmarking Multimodal LLMs on Recognition and Understanding over Chemical Tables

Can Recommender Systems Teach Themselves? A Recursive Self-Improving Framework with Fidelity Control

ReNF: Rethinking the Design Space of Neural Long-Term Time Series Forecasters

LIHE: Linguistic Instance-Split Hyperbolic-Euclidean Framework for Weakly-Supervised Generalized Referring Expression Comprehension

Unify in Isolation: Unified Data Synthesis for Divergent Multi-Stage Systems

Nested Hash Layer: A Plug-and-play Module for Multiple-length Hash Code Learning

Accelerating Large Language Model Inference via Speculative Decoding with Progressive Tree Drafting

Discovering and Leveraging Entropy-Complexity Relationships for Efficient Large Language Model Reasoning

A2R: An Asymmetric Two-Stage Reasoning Framework for Parallel Reasoning

MARCOS: Deep Thinking by Markov Chain of Continuous Thoughts

LogicTree: Improving Complex Reasoning of LLMs via Instantiated Multi-step Synthetic Logical Data

What Makes In-context Learning Effective for Mathematical Reasoning

Improving Time Series Forecasting via Instance-aware Post-hoc Revision

P-Law: Predicting Quantitative Scaling Law with Entropy Guidance in Large Recommendation Models

CogMath: Assessing LLMs' Authentic Mathematical Ability from a Human Cognitive Perspective

Accurate KV Cache Eviction via Anchor Direction Projection for Efficient LLM Inference

ToolACE: Winning the Points of LLM Function Calling

HyperTree Planning: Enhancing LLM Reasoning via Hierarchical Thinking

Unveiling the Magic of Code Reasoning through Hypothesis Decomposition and Amendment

TimeCapsule: Solving the Jigsaw Puzzle of Long-Term Time Series Forecasting with Compressed Predictive Representations

Multi-play Multi-armed Bandit Model with Scarce Sharable Arm Capacities

From Feature Interaction to Feature Generation: A Generative Paradigm of CTR Prediction Models

Diffusion Auto-regressive Transformer for Effective Self-supervised Time Series Forecasting

Harnessing Query Heterogeneity for Cost-Effective Proactive Caching in LLM Inference

LightSAM: Parameter-Agnostic Sharpness-Aware Minimization

Learn while Unlearn: An Iterative Unlearning Framework for Generative Language Models

TestAgent: An Adaptive and Intelligent Expert for Human Assessment

Multiple-play Stochastic Bandits with Prioritized Resource Sharing

Strengthening Federated Learning: Surrogate Data-Guided Aggregation for Robust Backdoor Defense

CogMath: Evaluating LLMs' Authentic Mathematical Ability from a Cognitive Perspective

Beyond Layers: A Global Message-Passing Mechanism for Heterophilic Graphs