Joseph E. Gonzalez

Associate Professor@University of California, Berkeley·美国·OpenReview

研究方向

Agentic Systems · LLM Evaluation · Training Large Language Models · Neural Network Design for Efficient Computer Vision · Autonomous Driving Systems · Distributed Machine Learning · Systems for Machine Learning · Inference Systems · Probabilistic Inference · Robotic Systems · Explainable AI · Inference In Probabilistic Graphical Models · Graph Processing Systems · Gaussian Process Models

Joseph E. Gonzalez

Search Arena: Analyzing Search-Augmented LLMs

vCache: Verified Semantic Prompt Caching

vAttention: Verified Sparse Attention via Sampling

SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse–Linear Attention

MFCL: A Multi-modal Function Calling Evaluation for Large Language Models

Are Large Reasoning Models Interruptible?

BARE: Leveraging Base Language Models for Few-Shot Synthetic Data Generation

Shepherd: Pattern-Guided Trajectory Selection for Coding Agents on SWE-Bench

Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling

HashAttention: Semantic Sparsity for Faster Inference

SimpleStrat: Diversifying Language Model Generation with Stratification

SuperCorrect: Advancing Small LLM Reasoning with Thought Template Distillation and Self-Correction

RouteLLM: Learning to Route LLMs from Preference Data

How to Evaluate Reward Models for RLHF

From Crowdsourced Data to High-quality Benchmarks: Arena-Hard and Benchbuilder Pipeline

Visual Haystacks: A Vision-Centric Needle-In-A-Haystack Benchmark

Bench-O-Matic: Automating Benchmark Curation from Crowdsourced Data

VibeCheck: Discover and Quantify Qualitative Differences in Large Language Models

Post-Training Sparse Attention with Double Sparsity

The Berkeley Function Calling Leaderboard (BFCL): From Tool Use to Agentic Evaluation of Large Language Models

SimpleStrat: Diversifying Language Model Generation with Stratification

Test-Time RAG: Enhancing Long Context Understanding in LLMs with Retrieval-Augmented Mechanisms