Han Wang
@HanWang98
Followers
243
Following
807
Media
6
Statuses
134
PhD student @unc @unccs @unc_ai_group; Intern @AMD; Formerly @AmazonScience @MSFTResearch @NlpWestlake. RT & like ≠ endorsements. Views are my own. He/him
Chapel Hill, NC, USA
Joined July 2019
🚨Real-world retrieval is messy: queries can be ambiguous, or documents may conflict/have incorrect/irrelevant info. How can we jointly address all these problems?. We introduce:.➡️ RAMDocs, a challenging dataset with ambiguity, misinformation, and noise. ➡️ MADAM-RAG, a
2
33
69
RT @meetdavidwan: 🎉 Our paper, GenerationPrograms, which proposes a modular framework for attributable text generation, has been accepted t….
0
21
0
RT @ArchikiPrasad: 🎉 Glad to see our work on handling conflicting & noisy evidence and ambiguous queries in RAG systems (via a new benchmar….
0
12
0
🥳 Excited to share our work -- Retrieval-Augmented Generation with Conflicting Evidence -- on addressing conflict in RAG due to ambiguity, misinformation, and noisy/irrelevant evidence has been accepted to @COLM_conf #COLM2025!. Our new benchmark RAMDocs proves challenging for.
🚨Real-world retrieval is messy: queries can be ambiguous, or documents may conflict/have incorrect/irrelevant info. How can we jointly address all these problems?. We introduce:.➡️ RAMDocs, a challenging dataset with ambiguity, misinformation, and noise. ➡️ MADAM-RAG, a
0
18
37
RT @ZiyangW00: 🚨Introducing Video-RTS: Resource-Efficient RL for Video Reasoning with Adaptive Video TTS! . While RL-based video reasoning….
0
26
0
RT @hyunji_amy_lee: 🥳Excited to share that I’ll be joining @unccs as postdoc this fall. Looking forward to work with @mohitban47 & amazing….
0
26
0
RT @RenZhongzheng: 🥳 Excited to share that I’ll be joining the CS Department at UNC-Chapel Hill (@unccs @unc_ai_group) as an Assistant Prof….
0
14
0
RT @EliasEskin: 🎉 Excited to share that CAPTURe has been accepted to #ICCV2025! CAPTURe is a new benchmark for VLM reasoning that requires….
0
14
0
RT @shoubin621: 🎉Excited to announce VEEGIE has been accepted to #ICCV2025 ! VEGGIE is a unified MLLM + Diffusion framework for instruction….
0
17
0
RT @shoubin621: New paper Alert 🚨 Introducing MEXA: A general and training-free multimodal reasoning framework via dynamic multi-expert ski….
0
27
0
RT @hyunji_amy_lee: 🚨 Want models to better utilize and ground on the provided knowledge? We introduce Context-INformed Grounding Supervisi….
0
41
0
RT @meetdavidwan: Excited to share GenerationPrograms! 🚀. How do we get LLMs to cite their sources? GenerationPrograms is attributable by d….
0
38
0
RT @meetdavidwan: Thanks for the discovering + sharing our work on contextualized late-interaction based multimodal content retrieval, Omar….
0
6
0
RT @lateinteraction: Wow I missed this extra fancy ColBERT model. > A late-interaction retriever which jointly encodes/contextualizes info….
0
18
0
RT @ZiyangW00: Excited to present VideoTree🌲 at #CVPR2025 Fri at 10:30AM!. VideoTree improves long-video QA via smart sampling:.-Query-adap….
0
19
0
RT @jmin__cho: Introducing CLaMR -- a late-interaction retriever for complex multimodal video content! 📽️📚. ➡️ Jointly encodes frames, spee….
0
9
0
How can a multimodal retriever accurately retrieve docs from massive online video content that spans multiple modalities?. We introduce CLaMR, a contextualized late-interaction retriever that jointly encodes all modalities and dynamically selects those containing the relevant.
Excited to share our new work, CLaMR! 🚀. We tackle multimodal content retrieval by jointly considering video, speech, OCR, and metadata. CLaMR learns to dynamically pick the right modality for your query, boosting retrieval by 25 nDCG@10 over single modality retrieval!. 🧐
1
9
18
RT @EliasEskin: Excited to announce CLaMR, our new retriever for multimodal documents! Strong performance improvements (+25 nDGC@10) compar….
0
10
0
RT @meetdavidwan: Excited to share our new work, CLaMR! 🚀. We tackle multimodal content retrieval by jointly considering video, speech, OCR….
0
62
0
RT @danadaeun: Excited to share Video-Skill-CoT🎬🛠️– a new framework for domain-adaptive video reasoning with skill-aware Chain-of-Thought (….
0
28
0