Papers by Haowei Fu

4 papers
HERMES: KV Cache as Hierarchical Memory for Efficient Streaming Video Understanding (2026.acl-long)

Copied to clipboard

Challenge: Existing models struggle to maintain stable understanding performance and low GPU memory overhead.
Approach: They propose a training-free architecture for real-time and accurate understanding of video streams . HERMES reuses a compact KV cache, enabling efficient streaming understanding .
Outcome: The proposed architecture achieves 10 faster TTFT compared to prior SOTA.
Ensemble Privacy Defense for Knowledge-Intensive LLMs against Membership Inference Attacks (2026.findings-eacl)

Copied to clipboard

Challenge: Large language models (LLMs) are the foundation of modern natural language processing, powering applications across diverse domains.
Approach: They propose a model-agnostic defense framework which aggregates and evaluates the outputs of a knowledge-injected LLM, a base LLM and a dedicated judge model to enhance resistance against membership inference attacks.
Outcome: The proposed framework reduces MIA success by up to 27.8% for SFT and 526.3% for RAG compared to inference-time baseline while maintaining answer quality.
Risk-Controlled Event-Driven Cascading Updates for Knowledge Graph Consistency Restoration (2026.findings-acl)

Copied to clipboard

Challenge: Knowledge Graphs (KGs) typically treat updates as independent facts . factual, localized updates can contradict and invalidate previously correct knowledge .
Approach: They propose a model-agnostic framework for cascading KG update identification that leverages conformal prediction to provide reliable uncertainty guarantees over the cascade as a whole.
Outcome: The proposed framework provides reliable uncertainty guarantees over the cascade as a whole . it integrates large language models to enrich event representations with world knowledge.
CoreCodeBench: Decoupling Code Intelligence via Fine-Grained Repository-Level Tasks (2026.acl-long)

Copied to clipboard

Challenge: Existing large language models for software engineering rely on coarse-grained pass rates obscuring specific cognitive bottlenecks.
Approach: They propose a repository-level benchmark that dissects coding capabilities through atomized tasks.
Outcome: The proposed framework achieves a 78.55% validity yield, surpassing the 31.7% retention rate of SWE-bench-Verified.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations