7 papers
DEdit: Iterative Draft Editing for Speculative Decoding
Longxuan Yu, Bingsen Chen, Peng Shi +9
Speculative decoding accelerates autoregressive LLMs by having a lightweight drafter propose tokens that the target model verifies in parallel. Diffusion-based drafters further red…
Sequential Beats Joint: On the Interplay between On-Policy Distillation and RLVR
Boyan Li, Bingsen Chen, Chenghao Yang +3
Reinforcement learning with verifiable rewards (RLVR) and on-policy distillation (OPD) have emerged as two dominant methods for post-training reasoning LLMs. Prior work uses OPD's…
Small Language Models as Judges for Rubric-Based Reinforcement Learning
Fengyu Xie, Yilun Zhao, Bingsen Chen +2
Rubric-based reinforcement learning extends RL beyond tasks with exact answers or rule-based verifiers by scoring responses against instance-specific criteria. However, this makes…
LLaDA2.1: Speeding Up Text Diffusion via Token Editing
Tiwei Bie, Maosong Cao, Xiang Cao +47
While LLaDA2.0 showcased the scaling potential of 100B-level block-diffusion models and their inherent parallelization, the delicate equilibrium between decoding speed and generati…
Beyond Single-shot Writing: Deep Research Agents are Unreliable at Multi-turn Report Revision
Bingsen Chen, Boyan Li, Ping Nie +3
Existing benchmarks for Deep Research Agents (DRAs) treat report generation as a single-shot writing task, which fundamentally diverges from how human researchers iteratively draft…
Inter-Passage Verification for Multi-evidence Multi-answer QA
Bingsen Chen, Shengjie Wang, Xi Ye +1
Multi-answer question answering (QA), where questions can have many valid answers, presents a significant challenge for existing retrieval-augmented generation-based QA systems, as…
Arithmetic Reasoning with LLM: Prolog Generation & Permutation
Xiaocheng Yang, Bingsen Chen, Yik-Cheung Tam
Instructing large language models (LLMs) to solve elementary school math problems has shown great success using Chain of Thought (CoT). However, the CoT approach relies on an LLM t…