Papers by Yifei Cao

12 papers
Muse: Towards Reproducible Long-Form Song Generation with Fine-Grained Style Control (2026.findings-acl)

Copied to clipboard

Challenge: Recent commercial systems such as Suno demonstrate strong capabilities in long-form song generation, but academic research remains non-reproducible due to the lack of publicly available training data.
Approach: They propose a system for long-form song generation with fine-grained style conditioning that includes a licensed synthetic dataset and a song generation model, Muse.
Outcome: The proposed system achieves competitive performance on phoneme error rate, text–music style similarity, and audio aesthetic quality while enabling controllable segment-level generation across different musical structures.
Head-wise Shareable Attention for Large Language Models (2024.findings-emnlp)

Copied to clipboard

Challenge: Large Language Models (LLMs) suffer from huge number of parameters, which restricts their deployment on edge devices.
Approach: They propose two methods that share parameters across attention heads to reduce memory usage and reduce performance drop by using coarse-grained weight sharing rules.
Outcome: The proposed methods reuse pre-trained weights without retraining and then share, denoted as PostShare.
VPTQ: Extreme Low-bit Vector Post-Training Quantization for Large Language Models (2024.emnlp-main)

Copied to clipboard

Challenge: Recent research has focused on pushing weight-only quantization to extremely low-bit due to numerical representation limitations.
Approach: They propose a vector-based quantization approach that pushes LLMs to extremely low-bit . they propose scalar-based weight quantization that reduces memory requirements and optimizes storage costs .
Outcome: The proposed method reduces model quantization perplexity by 0.01-0.34 on LLaMA-2, 0.38-0.68 on mistral-7B, 4.41-7.34, on llaMA-3 on QA tasks on average.
Topology Matters: Measuring Memory Leakage in Multi-Agent LLMs (2026.findings-acl)

Copied to clipboard

Challenge: Multi-agent LLMs are rapidly moving from prototype to real-world use . network topology is a first-order security parameter in multi-aggent systems .
Approach: They propose a framework for comparing topology-conditioned memory leakage in multi-agent LLM systems.
Outcome: The proposed framework evaluates topology-conditioned memory leakage in multi-agent LLM systems.
PGPO: Enhancing Agent Reasoning via Pseudocode-style Planning Guided Preference Optimization (2025.findings-acl)

Copied to clipboard

Challenge: Existing LLM agents generate verbose and inefficient natural language plans to guide reasoning, which restricts agents’ ability to generalize across similar tasks.
Approach: They propose a pseudocode-style planning guide optimization method that captures the structural logic of reasoning and uses two planning-oriented rewards to enhance agent learning.
Outcome: The proposed method outperforms existing LLM agents on representative agent benchmarks and outperformed the current leading baselines.
LaCo: Large Language Model Pruning via Layer Collapse (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for model quantization, knowledge distillation, and model pruning are limited by hardware support limitations and the need for extensive training.
Approach: They propose a layer-wise structured pruner that collapses rear model layers into a prior layer and enables a rapid reduction in model size while preserving the model structure.
Outcome: The proposed pruner outperforms state-of-the-art pruning methods at pruning ratios of 25-30% and maintains an average task performance of over 80% at different pruning ratio.
MAGRET: Machine-generated Text Detection with Rewritten Texts (2025.coling-main)

Copied to clipboard

Challenge: Existing studies focus on detecting machine-generated text in open-source models, but their performance on closed-source large models is limited.
Approach: They propose a method to detect rewritten text from large language models using a BERT encoder and propose to refine it to achieve semantic alignment.
Outcome: The proposed method outperforms baseline methods on three text-generated datasets.
TOOL-ED: Enhancing Empathetic Response Generation with the Tool Calling Capability of LLM (2025.coling-main)

Copied to clipboard

Challenge: Empathetic conversation is a crucial characteristic in daily conversations between individuals.
Approach: They propose an Emotional Knowledge Tool Calling framework which encapsulates commonsense knowledge bases as empathetic tools, enabling LLMs to integrate external knowledge flexibly.
Outcome: The proposed framework can generate empathetic responses effectively on the TOOL-ED dataset.
To See a World in a Spark of Neuron: Disentangling Multi-Task Interference for Training-Free Model Merging (2025.emnlp-main)

Copied to clipboard

Challenge: Existing approaches to model merging ignore the fundamental roles of neurons, connectivity and activation.
Approach: They propose a framework that relies on neuronal mechanisms to mitigate task interference . they decomposed task-specific representations into two complementary subspaces . their results offer new insights into mitigating task interference and improving knowledge fusion .
Outcome: The proposed framework reduces task interference within neurons and improves knowledge fusion.
LESA: Learnable LLM Layer Scaling-Up (2025.acl-long)

Copied to clipboard

Challenge: Existing methods for depth scaling-up rely on empirical heuristic rules for layer duplication, resulting in poor initialization and slower convergence during continual pre-training.
Approach: They propose a method for learning latent parameters between layers by concatenating parameters from each layer and applying Singular Value Decomposition.
Outcome: Experiments show that LESA outperforms baseline models with less than half the cost of existing methods.
From Scores to Preferences: Redefining Evaluation Paradigm for Speech Quality Reward Modeling (2026.findings-acl)

Copied to clipboard

Challenge: Experimental results show that the MOS-aware GRM significantly improves fine-grained speech quality discrimination.
Approach: They propose a MOS-aware reward model that incorporates MOS gap into reward function during reinforcement learning.
Outcome: The proposed model significantly improves fine-grained speech quality discrimination.
Beyond Scaling: Measuring and Predicting the Upper Bound of Knowledge Retention in Language Model Pre-Training (2026.acl-long)

Copied to clipboard

Challenge: Existing methods to predict performance of large language models are lacking . authors propose a size-dependent mutual information predictor for closed-book question answering accuracy .
Approach: They propose a size-dependent mutual information predictor that integrates knowledge frequency, knowledge specificity, and model size to forecast closed-book question answering accuracy.
Outcome: The proposed method outperforms baseline models and achieves R2 > 0.7 in predicting QA accuracy without additional training.

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