5 citations · 16 across the 38 of their papers we have counts for
42 papers
Shaping Latent Mixture-of-Experts: From Expert Geometry to Layerwise Capacity Allocation
Yaoxiang Wang, Kai Zhang, Ruobing Xie +4
Fine-grained mixture-of-experts (MoE) models commonly increase expert count by narrowing each expert, treating expert shape as a by-product of routing granularity. LatentMoE moves…
Efficient Reasoning Exploration via State-Conditioned Latent Steering with Progress Guidance
Hengyuan Zhang, Chenming Shang, Zunhai Su +10
Best-of- is a widely used inference strategy for complex reasoning, whose effectiveness depends on whether sampled candidates can cover diverse and high-quality reasoning paths.…
Higher-Dimensional Rotary Position Embedding
Yixing Li, Ruobing Xie, Yudong Zhang +3
Transformers rely on position embedding mechanisms in long context modeling in most cases. Rotary Position Embedding (RoPE) embeds positional information with independent 2D rotati…
Hybrid Gated Attention
Zekun Zhou, Ruobing Xie, Lanrui Wang +1
Gated attention is an effective approach to mitigate attention sinks and enhance the representational capacity of attention. To further extend its effectiveness-efficiency Pareto f…
AngelSpec: Towards Real-World High Performance Inference with Speculative Decoding
Hong Liu, Rui Cen, Junhan Shi +10
Speculative decoding accelerates large language model inference without changing the target distribution, but no single drafting structure performs best across real-world workloads…
FastMix: Fast Data Mixture Optimization via Gradient Descent
Haoru Tan, Sitong Wu, Yanfeng Chen +5
While large and diverse datasets have driven recent advances in large models, identifying the optimal data mixture for pre-training and post-training remains a significant open pro…
Beyond Encoder Accumulation: Measuring Encoder Roles in Multi-Encoder VLMs
Wei Ding, Yudong Zhang, Ruobing Xie +3
As foundation models scale toward fusing more heterogeneous visual streams, understanding how diverse encoders interact under joint training becomes a prerequisite for principled d…
Negative Advantages Is a Double-Edged Sword: Calibrating advantages in GRPO for Search Agents
Jiayi Wu, Ruobing Xie, Zeqian Huang +6
Search agents achieve strong question-answering performance through multi-turn interactions with search engines, with Group Relative Policy Optimization (GRPO) being a widely used…
MHSA: A Lightweight Framework for Mitigating Hallucinations via Steered Attention in LVLMs
Wei Ding, Yilin Li, Yudong Zhang +4
Large vision-language models (LVLMs) have achieved remarkable performance across diverse multimodal tasks, yet they continue to suffer from hallucinations, generating content that…
Self-Distillation for Multi-Token Prediction
Guoliang Zhao, Ruobing Xie, An Wang +3
As Large Language Models (LLMs) scale up, inference efficiency becomes a critical bottleneck. Multi-Token Prediction (MTP) could accelerate LLM inference by predicting multiple fut…
Beyond Outliers: A Data-Free Layer-wise Mixed-Precision Quantization Approach Driven by Numerical and Structural Dual-Sensitivity
Hengyuan Zhang, Xinrong Chen, Zunhai Su +10
Layer-wise mixed-precision quantization (LMPQ) enables effective compression under extreme low-bit settings by allocating higher precision to sensitive layers. However, existing me…
LaSeR: Reinforcement Learning with Last-Token Self-Rewarding
Wenkai Yang, Weijie Liu, Ruobing Xie +4
Reinforcement Learning with Verifiable Rewards (RLVR) has recently emerged as a core paradigm for enhancing the reasoning capabilities of Large Language Models (LLMs). To address t…
PT-LLM: Post-Training Ternarization for Large Language Models
Xianglong Yan, Chengzhu Bao, Zhiteng Li +6
Large Language Models (LLMs) have shown impressive capabilities across diverse tasks, but their large memory and compute demands hinder deployment. Ternarization has gained attenti…
Towards a Comprehensive Scaling Law of Mixture-of-Experts
Guoliang Zhao, Yuhan Fu, Shuaipeng Li +10
Mixture-of-Experts (MoE) models have become the consensus approach for enabling parameter-efficient scaling and cost-effective deployment in large language models. However, existin…
Hybrid-Tower: Fine-grained Pseudo-query Interaction and Generation for Text-to-Video Retrieval
Bangxiang Lan, Ruobing Xie, Ruixiang Zhao +4
The Text-to-Video Retrieval (T2VR) task aims to retrieve unlabeled videos by textual queries with the same semantic meanings. Recent CLIP-based approaches have explored two framewo…
Proximal Supervised Fine-Tuning
Wenhong Zhu, Ruobing Xie, Rui Wang +3
Supervised fine-tuning (SFT) of foundation models often leads to poor generalization, where prior capabilities deteriorate after tuning on new tasks or domains. Inspired by trust-r…
Flexible Realignment of Language Models
Wenhong Zhu, Ruobing Xie, Weinan Zhang +1
Realignment becomes necessary when a language model (LM) fails to meet expected performance. We propose a flexible realignment framework that supports quantitative control of align…
Fighting Fire with Fire (F3): A Training-free and Efficient Visual Adversarial Example Purification Method in LVLMs
Yudong Zhang, Ruobing Xie, Yiqing Huang +5
Recent advances in large vision-language models (LVLMs) have showcased their remarkable capabilities across a wide range of multimodal vision-language tasks. However, these models…
The Security Threat of Compressed Projectors in Large Vision-Language Models
Yudong Zhang, Ruobing Xie, Xingwu Sun +4
The choice of a suitable visual language projector (VLP) is critical to the successful training of large visual language models (LVLMs). Mainstream VLPs can be broadly categorized…
The Climb Carves Wisdom Deeper Than the Summit: On the Noisy Rewards in Learning to Reason
Ang Lv, Ruobing Xie, Xingwu Sun +2
Recent studies on post-training large language models (LLMs) for reasoning through reinforcement learning (RL) typically focus on tasks that can be accurately verified and rewarded…
Hunyuan-TurboS: Advancing Large Language Models through Mamba-Transformer Synergy and Adaptive Chain-of-Thought
Tencent Hunyuan Team, Ao Liu, Botong Zhou +248
As Large Language Models (LLMs) rapidly advance, we introduce Hunyuan-TurboS, a novel large hybrid Transformer-Mamba Mixture of Experts (MoE) model. It synergistically combines Mam…
QAVA: Query-Agnostic Visual Attack to Large Vision-Language Models
Yudong Zhang, Ruobing Xie, Jiansheng Chen +3
In typical multimodal tasks, such as Visual Question Answering (VQA), adversarial attacks targeting a specific image and question can lead large vision-language models (LVLMs) to p…
Large Language Model Empowered Recommendation Meets All-domain Continual Pre-Training
Haokai Ma, Yunshan Ma, Ruobing Xie +5
Recent research efforts have investigated how to integrate Large Language Models (LLMs) into recommendation, capitalizing on their semantic comprehension and open-world knowledge f…
TransMamba: A Sequence-Level Hybrid Transformer-Mamba Language Model
Yixing Li, Ruobing Xie, Zhen Yang +8
Transformers are the cornerstone of modern large language models, but their quadratic computational complexity limits efficiency in long-sequence processing. Recent advancements in…