Papers by Zhaozhuo Xu

19 papers
ScaleLLM: A Resource-Frugal LLM Serving Framework by Optimizing End-to-End Efficiency (2024.emnlp-industry)

Copied to clipboard

Challenge: Large language models (LLMs) are widely used in commercial applications . low latency is crucial due to system latency, query concurrency, and computational resources constraints.
Approach: They propose a system that can be resource-efficiently served by addressing bottlenecks beyond LLM inference . they propose 4.3 speed up over vLLM and 1.5 higher throughput .
Outcome: The proposed system outperforms state-of-the-arts with 1.5 higher throughput . it achieves 4.3 speed up with 64 concurrent requests on Mixtral 8x7B .
On Efficient Retrieval of Top Similarity Vectors (D19-1)

Copied to clipboard

Challenge: Existing representation learning methods such as Word2vec represent word embeddings in the semantic space.
Approach: They propose an efficient method for searching vectors via a non-metric matching function: inner product.
Outcome: Experiments on data representations learned for different machine learning tasks show the proposed method outperforms existing methods.
Word Salad Chopper: Reasoning Models Waste A Ton Of Decoding Budget On Useless Repetitions, Self-Knowingly (2025.emnlp-main)

Copied to clipboard

Challenge: Large Reasoning Models (LRMs) are often bottlenecked by the high cost of output tokens.
Approach: They propose a lightweight, turnkey component for Large Reasoning Models that is minimally invasive to its reasoning trajectory.
Outcome: The proposed component is lightweight and low overhead, and lacks semantic value.
TensorOpera Router: A Multi-Model Router for Efficient LLM Inference (2024.emnlp-industry)

Copied to clipboard

Challenge: Large Language Models (LLMs) have demonstrated remarkable performance across a diverse set of domain-specific tasks.
Approach: They propose a non-monolithic LLM querying system that seamlessly integrates various LLM experts into a single query interface and dynamically routes incoming queries to the most high-performant expert based on query’s requirements.
Outcome: The proposed model improves query efficiency by 40% and costs by 30% while maintaining or enhancing model performance by 10%.
In Defense of Structural Sparse Adapters for Concurrent LLM Serving (2024.findings-emnlp)

Copied to clipboard

Challenge: Large language models (LLMs) require adapters to fine tune performance without extensive retraining.
Approach: They propose a system that uses structurally sparse adapters to serve LLMs with multiple structurally-sparse axons.
Outcome: The proposed system achieves 2.12 speedup over low-rank adapters on 96 adapters with a single GPU.
DEL-ToM: Inference-Time Scaling for Theory-of-Mind Reasoning via Dynamic Epistemic Logic (2025.emnlp-main)

Copied to clipboard

Challenge: Theory-of-Mind (ToM) tasks pose a unique challenge for large language models (LLMs), which often lack the capability for dynamic logical reasoning.
Approach: They propose a framework that decomposes ToM tasks into a sequence of belief updates grounded in Dynamic Epistemic Logic (DEL) they use data generated automatically via a DEL simulator to train a verifier, which is called the Process Belief Model (PBM).
Outcome: The proposed framework improves verifiable ToM reasoning through inference-time scaling rather than architectural changes.
KV Cache Compression, But What Must We Give in Return? A Comprehensive Benchmark of Long Context Capable Approaches (2024.findings-emnlp)

Copied to clipboard

Challenge: Long context capability is a crucial competency for large language models as it mitigates the human struggle to digest long-form texts.
Approach: They propose to evaluate 10+ state-of-the-art approaches for long context-capable LLMs.
Outcome: The proposed methods are compared against 10+ state-of-the-art approaches across seven categories of long context tasks.
Token-wise Influential Training Data Retrieval for Large Language Models (2024.acl-long)

Copied to clipboard

Challenge: Large language models (LLMs) have been widely used in various industries due to their unprecedented scale and impressive capabilities derived from the massive training dataset.
Approach: They propose a framework that can estimate the influence of training data by caching and retrieval.
Outcome: The proposed framework can estimate the influence of training data within minutes, achieving over a speedup of 6,326x.
LLMs and Copyright Risks: Benchmarks and Mitigation Approaches (2025.naacl-tutorial)

Copied to clipboard

Challenge: Large Language Models (LLMs) have revolutionized natural language processing, but their widespread use has raised significant copyright concerns.
Approach: This tutorial will provide an overview of relevant copyright principles and their application to AI and examine specific copyright issues in LLM development and deployment.
Outcome: The course will provide an overview of relevant copyright principles and their application to AI, followed by an examination of specific copyright issues in LLM development and deployment.
ALinFiK: Learning to Approximate Linearized Future Influence Kernel for Scalable Third-Parity LLM Data Valuation (2025.naacl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) heavily rely on high-quality training data, making data valuation crucial for optimizing model performance.
Approach: They propose a third-party data valuation approach that assesses the value of individual data samples and proposes a learning strategy to approximate LinFiK.
Outcome: The proposed approach surpasses baselines in effectiveness and efficiency, showing significant scalability advantages as LLM parameters increase.
Copyright Detective: A Forensic System to Evidence LLMs Flickering Copyright Leakage Risks (2026.acl-demo)

Copied to clipboard

Challenge: **Copyright Detective** is the first interactive forensic system for detecting, analyzing, and visualizing potential copyright risks in LLM outputs.
Approach: They propose a system that detects copyright infringements and visualizes them . they use content recall testing, paraphrase-level similarity analysis and persuasive jailbreak probing .
Outcome: The proposed system detects, analyzes, and visualizes potential copyright risks in LLM outputs.
Collision to Cognition: Hash-Driven Graph Construction for Efficient RAG (2026.acl-long)

Copied to clipboard

Challenge: Retrieval-augmented generation (RAG) has been used for enhancing large language models with external knowledge.
Approach: They propose a framework for mining efficient graph structures via hashing to enhance RAG . they adopt an inductive paradigm where global graph structure emerges from local hash collisions .
Outcome: The proposed framework outperforms existing baselines while requiring no GPU resources or token budget.
Taming Language Models for Text-attributed Graph Learning with Decoupled Aggregation (2025.acl-long)

Copied to clipboard

Challenge: Existing approaches to learning text-attributed graphs neglect interaction between textual and structural information.
Approach: They propose a framework that integrates textual and structural information into TAG learning . they propose combining semantic aggregation and structural aggregations to improve learning a .
Outcome: The proposed framework outperforms state-of-the-art learning methods while requiring less resources.
Query-Aware Knowledge Retrieval via Hyperbolic Structuring (2026.acl-long)

Copied to clipboard

Challenge: Existing approaches focus primarily on retrieving isolated factual knowledge entities while neglecting the critical reasoning relationships.
Approach: They propose a query-centric retrieval framework that explicitly integrates structured knowledge graphs to support complex reasoning tasks.
Outcome: Extensive experiments on three benchmark datasets show that HyperRAG outperforms baselines.
Rescorla-Wagner Steering of LLMs for Undesired Behaviors over Disproportionate Inappropriate Context (2025.emnlp-main)

Copied to clipboard

Challenge: Incorporating external context can enhance the response quality of Large Language Models (LLMs). however, real-world contexts often mix relevant information with disproportionate inappropriate content.
Approach: They propose a Poisoned Context Testbed to pair queries with real-world contexts . they propose 'rw-Steering' to internalize inappropriate signals .
Outcome: The proposed model improves response quality by 39.8% and reverses undesirable behavior curve.
Do LLMs Know to Respect Copyright Notice? (2024.emnlp-main)

Copied to clipboard

Challenge: Existing studies have focused on the occurrence of copyright violations in LLM output, but a negative answer would suggest that LLMs will become the primary facilitator and accelerator of copy right infringement behavior.
Approach: They propose to examine whether LLMs respect copyright information in user input . they use a set of language models, user prompts, and copyrighted materials .
Outcome: The proposed model will be the primary facilitator and accelerator of copyright infringement behavior, the study finds . the study also provides a benchmark dataset serving as a test bed for evaluating infringement behaviors by LLMs .
Profiling LLM’s Copyright Infringement Risks under Adversarial Persuasive Prompting (2025.findings-emnlp)

Copied to clipboard

Challenge: Large Language Models have demonstrated impressive capabilities in text generation but raise concerns regarding potential copyright infringement.
Approach: They propose a structured persuasion workflow to analyze the influence of persuasive prompts on LLM outputs.
Outcome: The proposed method analyzes the influence of persuasive prompts on LLM outputs.
QUEST: Efficient Extreme Multi-Label Text Classification with Large Language Models on Commodity Hardware (2024.findings-emnlp)

Copied to clipboard

Challenge: Extreme multi-label text classification (EMTC) involves predicting multiple labels from a vast pool of candidates based on a user’s textual query.
Approach: They propose a Quantized and Efficient Learning with Sampling Technique that uses a hash sampling module to reduce the data volume to one-fourth of its original size.
Outcome: Extensive experiments show that QUEST outperforms existing methods while requiring fewer computational resources.
Structural Contrastive Representation Learning for Zero-shot Multi-label Text Classification (2022.findings-emnlp)

Copied to clipboard

Challenge: Existing approaches for zero-shot multi-label text classification struggle with accuracy and poor training efficiency.
Approach: They propose a structural contrastive representation learning approach that uses randomized text segmentation to generate high-quality contrastive pairs.
Outcome: The proposed approach improves accuracy and speed up training time on publicly available datasets.

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