9 citations · 19 across the 24 of their papers we have counts for
25 papers
Enough is as good as a feast: A Comprehensive Analysis of How Reinforcement Learning Mitigates Task Conflicts in LLMs
Zixuan Ren, Jinliang Lu, Junhong Wu +5
Model merging plays a crucial role in consolidating multiple specialized models into a single, unified model, especially in the era of large language models (LLMs). Recent research…
Sparse Growing Transformer: Training-Time Sparse Depth Allocation via Progressive Attention Looping
Yao Chen, Yilong Chen, Yinqi Yang +9
Existing approaches to increasing the effective depth of Transformers predominantly rely on parameter reuse, extending computation through recursive execution. Under this paradigm,…
KnowRL: Boosting LLM Reasoning via Reinforcement Learning with Minimal-Sufficient Knowledge Guidance
Linhao Yu, Tianmeng Yang, Siyu Ding +8
RLVR improves reasoning in large language models, but its effectiveness is often limited by severe reward sparsity on hard problems. Recent hint-based RL methods mitigate sparsity…
DreamKG: A KG-Augmented Conversational System for People Experiencing Homelessness
Javad M Alizadeh, Genhui Zheng, Chiu C Tan +7
People experiencing homelessness (PEH) face substantial barriers to accessing timely, accurate information about community services. DreamKG addresses this through a knowledge grap…
Learning to Generate via Understanding: Understanding-Driven Intrinsic Rewarding for Unified Multimodal Models
Jiadong Pan, Liang Li, Yuxin Peng +6
Recently, unified multimodal models (UMMs) have made remarkable progress in integrating visual understanding and generation, demonstrating strong potential for complex text-to-imag…
VideoAR: Autoregressive Video Generation via Next-Frame & Scale Prediction
Longbin Ji, Xiaoxiong Liu, Junyuan Shang +4
Recent advances in video generation have been dominated by diffusion and flow-matching models, which produce high-quality results but remain computationally intensive and difficult…
V-ITI: Mitigating Hallucinations in Multimodal Large Language Models via Visual Inference-Time Intervention
Nan Sun, Zhenyu Zhang, Xixun Lin +8
Multimodal Large Language Models (MLLMs) excel in numerous vision-language tasks yet suffer from hallucinations, producing content inconsistent with input visuals, that undermine r…
Blink: Dynamic Visual Token Resolution for Enhanced Multimodal Understanding
Yuchen Feng, Zhenyu Zhang, Naibin Gu +8
Multimodal large language models (MLLMs) have achieved remarkable progress on various vision-language tasks, yet their visual perception remains limited. Humans, in comparison, per…
Elastic MoE: Unlocking the Inference-Time Scalability of Mixture-of-Experts
Naibin Gu, Zhenyu Zhang, Yuchen Feng +8
Mixture-of-Experts (MoE) models typically fix the number of activated experts at both training and inference. However, real-world deployments often face heterogeneous hardware,…
Advantageous Parameter Expansion Training Makes Better Large Language Models
Naibin Gu, Yilong Chen, Zhenyu Zhang +7
Although scaling up the number of trainable parameters can effectively improve the training performance of large language models, it also leads to increased computational overhead.…
Inner Thinking Transformer: Leveraging Dynamic Depth Scaling to Foster Adaptive Internal Thinking
Yilong Chen, Junyuan Shang, Zhenyu Zhang +7
Large language models (LLMs) face inherent performance bottlenecks under parameter constraints, particularly in processing critical tokens that demand complex reasoning. Empirical…
BeamLoRA: Beam-Constraint Low-Rank Adaptation
Naibin Gu, Zhenyu Zhang, Xiyu Liu +7
Due to the demand for efficient fine-tuning of large language models, Low-Rank Adaptation (LoRA) has been widely adopted as one of the most effective parameter-efficient fine-tunin…
Curiosity-Driven Reinforcement Learning from Human Feedback
Haoran Sun, Yekun Chai, Shuohuan Wang +3
Reinforcement learning from human feedback (RLHF) has proven effective in aligning large language models (LLMs) with human preferences, but often at the cost of reduced output dive…
Mixture of Hidden-Dimensions Transformer
Yilong Chen, Junyuan Shang, Zhengyu Zhang +6
Transformer models encounter challenges in scaling hidden dimensions efficiently, as uniformly increasing them inflates computational and memory costs while failing to emphasize th…
Upcycling Instruction Tuning from Dense to Mixture-of-Experts via Parameter Merging
Tingfeng Hui, Zhenyu Zhang, Shuohuan Wang +3
Mixture-of-Experts (MoE) shines brightly in large language models (LLMs) and demonstrates outstanding performance in plentiful natural language processing tasks. However, existing…
MA-RLHF: Reinforcement Learning from Human Feedback with Macro Actions
Yekun Chai, Haoran Sun, Huang Fang +3
Reinforcement learning from human feedback (RLHF) has demonstrated effectiveness in aligning large language models (LLMs) with human preferences. However, token-level RLHF suffers…
NACL: A General and Effective KV Cache Eviction Framework for LLMs at Inference Time
Yilong Chen, Guoxia Wang, Junyuan Shang +7
Large Language Models (LLMs) have ignited an innovative surge of AI applications, marking a new era of exciting possibilities equipped with extended context windows. However, hosti…
HFT: Half Fine-Tuning for Large Language Models
Tingfeng Hui, Zhenyu Zhang, Shuohuan Wang +3
Large language models (LLMs) with one or more fine-tuning phases have become a necessary step to unlock various capabilities, enabling LLMs to follow natural language instructions…
Autoregressive Pre-Training on Pixels and Texts
Yekun Chai, Qingyi Liu, Jingwu Xiao +3
The integration of visual and textual information represents a promising direction in the advancement of language models. In this paper, we explore the dual modality of language--b…
On Training Data Influence of GPT Models
Yekun Chai, Qingyi Liu, Shuohuan Wang +3
Amidst the rapid advancements in generative language models, the investigation of how training data shapes the performance of GPT models is still emerging. This paper presents GPTf…
Tool-Augmented Reward Modeling
Lei Li, Yekun Chai, Shuohuan Wang +4
Reward modeling (a.k.a., preference modeling) is instrumental for aligning large language models with human preferences, particularly within the context of reinforcement learning f…
ERNIE-Music: Text-to-Waveform Music Generation with Diffusion Models
Pengfei Zhu, Chao Pang, Yekun Chai +5
In recent years, the burgeoning interest in diffusion models has led to significant advances in image and speech generation. Nevertheless, the direct synthesis of music waveforms f…
ERNIE 3.0 Tiny: Frustratingly Simple Method to Improve Task-Agnostic Distillation Generalization
Weixin Liu, Xuyi Chen, Jiaxiang Liu +4
Task-agnostic knowledge distillation attempts to address the problem of deploying large pretrained language model in resource-constrained scenarios by compressing a large pretraine…
ERNIE-SAT: Speech and Text Joint Pretraining for Cross-Lingual Multi-Speaker Text-to-Speech
Xiaoran Fan, Chao Pang, Tian Yuan +9
Speech representation learning has improved both speech understanding and speech synthesis tasks for single language. However, its ability in cross-lingual scenarios has not been e…