publications

2026

1. 

   CORRECT: COndensed eRror RECognition via knowledge Transfer in multi-agent systems

   Yifan Yu, Moyan Li, Shaoyuan Xu, and 4 more authors

   In International Conference on Machine Learning, 2026

   HTML

2. 

   XRPO: Pushing the limits of GRPO with Targeted Exploration and Exploitation

   Udbhav* Bamba, Minghao* Fang, Yifan* Yu, and 2 more authors

   In International Conference on Machine Learning, 2026

   HTML

3. 

   Introspective Diffusion Language Models

   Yifan* Yu, Yuqing* Jian, Junxiong Wang, and 12 more authors

   arXiv preprint arXiv:2604.11035, 2026

   Abs HTML

   Diffusion language models (DLMs) offer a compelling promise: parallel token generation could break the sequential bottleneck of autoregressive (AR) decoding. Yet in practice, DLMs consistently lag behind AR models in quality. We argue that this gap stems from a fundamental failure of introspective consistency: AR models agree with what they generate, whereas DLMs often do not. We introduce the Introspective Diffusion Language Model (I-DLM), which uses introspective strided decoding (ISD) to verify previously generated tokens while advancing new ones in the same forward pass. Empirically, I-DLM-8B is the first DLM to match the quality of its same-scale AR counterpart, outperforming LLaDA-2.1-mini (16B) by +26 on AIME-24 and +15 on LiveCodeBench-v6 with half the parameters, while delivering 2.9-4.1x throughput at high concurrency. With gated LoRA, ISD enables bit-for-bit lossless acceleration.

4. OPPO: Accelerating PPO-based RLHF via Pipeline Overlap

   Kaizhuo Yan, Yingjie Yu, Yifan Yu, and 2 more authors

   In International Conference on Learning Representations, 2026

   HTML

2025

1. 

   IC-Cache: Efficient Large Language Model Serving via In-context Caching

   Yifan* Yu, Yu* Gan, Nikhil Sarda, and 7 more authors

   In Proceedings of the ACM SIGOPS 31st Symposium on Operating Systems Principles, Lotte Hotel World, Seoul, Republic of Korea, 2025

   31st SOSP Abs DOI

   Acceptance rate: 17 percent

   Large language models (LLMs) have excelled in various applications, yet serving them at scale is challenging due to their substantial resource demands and high latency. Our real-world studies reveal that over 70% of user requests to LLMs have semantically similar counterparts, suggesting the potential for knowledge transfer among requests. However, naively caching and reusing past responses leads to a big quality drop.In this paper, we introduce IC-Cache, a caching system that enables live LLM capability augmentation to improve serving efficiency: by leveraging historical request-response pairs from larger models as in-context examples, IC-Cache empowers small LLMs to imitate and even exceed the compositional abilities (e.g., reasoning) of their larger counterparts, enabling selective offloading of requests to reduce cost and latency. Achieving this live augmentation at scale introduces intricate trade-offs between response quality, latency, and system throughput. For a new request, IC-Cache efficiently selects similar, high-utility examples to prepend them to the new request’s input. At scale, it adaptively routes requests across LLMs of varying capabilities, accounting for response quality and serving loads. IC-Cache employs a cost-aware cache replay mechanism that refines example quality offline to maximize online cache utility and efficiency. Evaluations on millions of realistic requests demonstrate that IC-Cache improves LLM serving throughput by 1.4–5.9x and reduces latency by 28–71% without hurting response quality.

2024

1. 

   Loftq: Lora-fine-tuning-aware quantization for large language models

   Yixiao* Li, Yifan* Yu, Chen Liang, and 4 more authors

   The Twelfth International Conference on Learning Representations, 2024

   Oral Presentation

   About 1 percent of the submissions are selected as oral presentation

2023

1. 

   Losparse: Structured compression of large language models based on low-rank and sparse approximation

   Yixiao* Li, Yifan* Yu, Qingru Zhang, and 4 more authors

   In International Conference on Machine Learning, 2023