Profiles2024-2026
Yiwei Wang
University of California, Merced
17Papers
2Last author
22Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers17
- Parameters vs. Context: Fine-Grained Control of Knowledge Reliance in Language ModelsAuthor 3 of 7ICLR 2026 · 47 citations
- Reward and Guidance through Rubrics: Promoting Exploration to Improve Multi-Domain ReasoningAuthor 3 of 9ICML 2026 · 19 citations
- HiddenGuard: Fine-Grained Safe Generation with Specialized Representation RouterAuthor 3 of 6ACL 2026 · 14 citations
- WavefrontDiffusion: Dynamic Decoding Schedule for Improved ReasoningAuthor 6 of 6ICLR 2026 · 5 citations
- Gated Differentiable Working Memory for Long-Context Language ModelingAuthor 3 of 10ACL 2026 · 4 citations
- VideoStir: Understanding Long Videos via Spatio-Temporally Structured and Intent-Aware RAGAuthor 3 of 6ACL 2026 · 2 citations
- ChainMPQ: Interleaved Text-Image Reasoning Chains for Mitigating Relation HallucinationsAuthor 2 of 3ICLR 2026 · 1 citation
- a1: Steep Test-time Scaling Law via Environment Augmented GenerationAuthor 3 of 8ACL Findings 2026
- ReCon: Active Defense against Large Vision-Language Model Jailbreaks via Reverse Safety Concept InjectionAuthor 2 of 4ACL Findings 2026
- Context-DPO: Aligning Language Models for Context-FaithfulnessAuthor 3 of 14ACL Findings 2025 · 2 citations
- Decoding by Contrasting Knowledge: Enhancing Large Language Model Confidence on Edited FactsAuthor 4 of 6ACL 2025
- Is Factuality Enhancement a Free Lunch For LLMs? Better Factuality Can Lead to Worse Context-FaithfulnessAuthor 3 of 8ICLR 2025
- Making Every Step Effective: Jailbreaking Large Vision-Language Models Through Hierarchical KV EqualizationAuthor 2 of 7EMNLP Findings 2025
- SemVink: Advancing VLMs' Semantic Understanding of Optical Illusions via Visual Global ThinkingAuthor 3 of 3EMNLP 2025
- SLANG: New Concept Comprehension of Large Language ModelsAuthor 3 of 5EMNLP 2024 · 7 citations
- LPNL: Scalable Link Prediction with Large Language ModelsAuthor 3 of 5ACL Findings 2024 · 6 citations
- Adaptive Token Biaser: Knowledge Editing via Biasing Key EntitiesAuthor 3 of 7EMNLP Findings 2024
