activity
20192026
most citedLlama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders

5 citations · 19 across the 37 of their papers we have counts for

collaborators
Showing cs.CLShow all

42 papers · 1 filter

cs.CL2026

Memory Decoder at Scale: A Pretrained, Parametric Long-Term Memory

Rubin Wei, Jiaqi Cao, Jiarui Wang +4

Decoder-only language models entangle long-term memory and reasoning in a single parameter set, making it difficult to scale memory capacity independently. Memory Decoder introduce…

cs.CL2026

When Does Generating More Help? Disentangling Fixed-Source Synthesis from Source Expansion in Synthetic Data Scaling

Xu Guo, Jian Tong, Zhihui Lu +1

Synthetic data can be scaled along two routes: Source Expansion (SE), which enlarges the source by adding seed materials or generators, and Fixed-Source Synthesis (FSS), which hold…

cs.CL2026

Synthetic Pre-Pre-Training Improves Language Model Robustness to Noisy Pre-Training Data

Xu Guo, Runyu Peng, Jian Tong +4

Large language models (LLMs) rely on web-scale corpora for pre-training. The noise inherent in these datasets tends to obscure meaningful patterns and ultimately degrade model perf…

cs.CL2026

Kernel-Smith: A Unified Recipe for Evolutionary Kernel Optimization

He Du, Qiming Ge, Jiakai Hu +18

We present Kernel-Smith, a framework for high-performance GPU kernel and operator generation that combines a stable evaluation-driven evolutionary agent with an evolution-oriented…

cs.CL2026

How to Fine-Tune a Reasoning Model? A Teacher-Student Cooperation Framework to Synthesize Student-Consistent SFT Data

Zixian Huang, Kaichen Yang, Xu Huang +6

A widely adopted strategy for model enhancement is to use synthetic data generated by a stronger model for supervised fine-tuning (SFT). However, for emerging reasoning models like…

cs.CL2026

Rethinking Multiple-Choice Questions for RLVR: Unlocking Potential via Distractor Design

Xu Guo, Qiming Ge, Jian Tong +8

Reinforcement Learning with Verifiable Rewards (RLVR) significantly enhances the reasoning capabilities of Large Language Models. When applied to RLVR, Multiple-Choice Questions (M…