Papers by Ruibo Chen

7 papers
HATA: Trainable and Hardware-Efficient Hash-Aware Top-k Attention for Scalable Large Model Inference (2025.findings-acl)

Copied to clipboard

Challenge: Existing top-k attention methods struggle to strike a balance between efficiency and accuracy.
Approach: They propose a top-k attention approach that integrates low-overhead techniques into the Top-k Attention process to achieve 7.2 speedup compared to vanilla full attention.
Outcome: The proposed approach achieves 7.2 speedup compared to current top-k attention methods while maintaining model accuracy.
UniSonate: A Unified Model for Speech, Music, and Sound Effect Generation with Text Instructions (2026.acl-long)

Copied to clipboard

Challenge: Generative audio modeling has been fragmented into specialized tasks such as text-to-speech (TTS), text- to-music (TTM), and text-ta (TTA) specialized models require reference audio for timbre cloning and strict phoneme alignment, whereas TTA models generate unstructured textures from open-ended captions.
Approach: They propose a unified flow-matching framework capable of synthesizing speech, music, sound effects . they propose 'token injection mechanism' that projects unstructured environmental sounds into structured temporal latent space .
Outcome: The proposed framework achieves state-of-the-art performance in instruction-based TTS and TTM while maintaining competitive fidelity in TTA.
Breaking the Stigma! Unobtrusively Probe Symptoms in Depression Disorder Diagnosis Dialogue (2025.findings-naacl)

Copied to clipboard

Challenge: Recent efforts to improve depression-diagnosis-oriented dialogue systems ignore the Stigma preventing users from open conversations about their struggles.
Approach: They propose a method to promote a sense of unobtrusiveness within the dialogue system and assessing depression disorder by probing symptoms.
Outcome: The proposed method improves on baselines including unobtrusiveness evaluation of dialogue content and diagnostic accuracy.
Your Vision-Language Model Itself Is a Strong Filter: Towards High-Quality Instruction Tuning with Data Selection (2024.findings-acl)

Copied to clipboard

Challenge: Existing data selection methods for instruction-following large language models rely on unreliable scores or use downstream tasks for selection.
Approach: They propose a method that utilizes the VLM itself as a filter to select high-quality instruction-tuning data.
Outcome: The proposed method can reach better results compared to full data settings with merely about 15% samples and can achieve superior performance against competitive baselines.
ChatMusician: Understanding and Generating Music Intrinsically with LLM (2024.findings-acl)

Copied to clipboard

Challenge: Despite LLMs' impressive capabilities in musical knowledge, music reasoning remains an unsolved task.
Approach: They propose an open-source large language model (LLM) that integrates intrinsic musical abilities into LLaMA2 and GPT-3.5.
Outcome: The proposed model can understand and generate music with a pure text tokenizer without external multi-modal neural structures or tokenizers.
MoLA: MoE LoRA with Layer-wise Expert Allocation (2025.findings-naacl)

Copied to clipboard

Challenge: Recent efforts to integrate low-rank adaptation (LoRA) with the Mixture-of-Experts (MoE) have achieved performance comparable to full-parameter fine-tuning by tuning much fewer parameters.
Approach: They propose a parameter-efficient MoE method for low-rank adaptation with the Mixture-of-Experts (MoE) they use layers of LoRA experts to allocate more LoRA expert to middle layers .
Outcome: The proposed method outperforms baseline models on six well-known NLP and commonsense QA benchmarks on LLAMA-2, Mistral, and Gemma.
Improved Unbiased Watermark for Large Language Models (2025.acl-long)

Copied to clipboard

Challenge: Unbiased watermarks allow to distinguish between text generated by humans and machines without causing distortion.
Approach: They introduce a family of unbiased, Multi-Channel-based watermarks that partition the language model into segments and promote token probabilities within a selected segment based on a watermark key.
Outcome: The proposed watermarks preserve the original distribution of the language model and offer significant improvements in detectability and robustness over existing unbiased watermark systems.

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