159 citations · 387 across the 84 of their papers we have counts for
18 papers · 1 filter
HarnessTTT: Learn to Optimize Harness at Test-Time for Scientific Discovery
Yingzi Ma, Yulong Cao, Nanxi Li +5
Scientific discovery with language models requires finding the best valid solution within a finite computational budget. We study how feedback from this search can improve the proc…
SimpleDesign: A Joint Model for Protein Sequence and Structure Codesign
Jiarui Lu, Yuyang Wang, Yizhe Zhang +4
Proteins are fundamental to biological processes, with their function determined by the complex interplay between the amino acid sequence and the three-dimensional structure. Devel…
LT2: Linear-Time Looped Transformers
Chunyuan Deng, Yizhe Zhang, Rui-Jie Zhu +4
Looped Transformers (LT) have emerged as a powerful architecture by iterating their layers multiple times before decoding the final token. However, pairing them with full attention…
Primal Generation, Dual Judgment: Self-Training from Test-Time Scaling
Yizhu Jiao, Ruixiang Zhang, Richard Bai +3
Code generation is typically trained in the primal space of programs: a model produces a candidate solution and receives sparse execution feedback, often a single pass/fail bit. Te…
DACA-GRPO: Denoising-Aware Credit Assignment for Reinforcement Learning in Diffusion Language Models
Amin Karimi Monsefi, Dominic Culver, Nikhil Bhendawade +4
Diffusion large language models are a compelling alternative to autoregressive models, yet existing RL methods for diffusion treat all denoising steps as equally important and rely…
Trajectory as the Teacher: Few-Step Discrete Flow Matching via Energy-Navigated Distillation
Amin Karimi Monsefi, Dominic Culver, Nikhil Bhendawade +3
Discrete flow matching generates text by iteratively transforming noise tokens into coherent language, but may require hundreds of forward passes. Distillation uses the multi-step…
Unmasking On-Policy Distillation: Where It Helps, Where It Hurts, and Why
Mohammadreza Armandpour, Fatih Ilhan, David Harrison +6
On-policy distillation offers dense, per-token supervision for training reasoning models; however, it remains unclear under which conditions this signal is beneficial and under whi…
Expert Divergence Learning for MoE-based Language Models
Jiaang Li, Haibin Chen, Langming Liu +9
The Mixture-of-Experts (MoE) architecture is a powerful technique for scaling language models, yet it often suffers from expert homogenization, where experts learn redundant functi…
SpiralFormer: Looped Transformers Can Learn Hierarchical Dependencies via Multi-Resolution Recursion
Chengting Yu, Xiaobo Shu, Yadao Wang +8
Recursive (looped) Transformers decouple computational depth from parameter depth by repeatedly applying shared layers, providing an explicit architectural primitive for iterative…
LaDi-RL: Latent Diffusion Reasoning Prevents Entropy Collapse in Reinforcement Learning
Haoqiang Kang, Yizhe Zhang, Nikki Lijing Kuang +2
Reinforcement learning has become a central paradigm for improving LLM reasoning, but most existing methods optimize policies over discrete token sequences. This creates a mismatch…
MeSH: Memory-as-State-Highways for Recursive Transformers
Chengting Yu, Xiaobo Shu, Yadao Wang +8
Recursive transformers reuse parameters and iterate over hidden states multiple times, decoupling compute depth from parameter depth. However, under matched compute, recursive mode…
LaDiR: Latent Diffusion Enhances LLMs for Text Reasoning
Haoqiang Kang, Yizhe Zhang, Nikki Lijing Kuang +4
Large Language Models (LLMs) demonstrate their reasoning ability through chain-of-thought (CoT) generation. However, LLM's autoregressive decoding may limit the ability to revisit…
Flexible Language Modeling in Continuous Space with Transformer-based Autoregressive Flows
Ruixiang Zhang, Shuangfei Zhai, Jiatao Gu +6
Autoregressive models have driven remarkable progress in language modeling. Their foundational reliance on discrete tokens, unidirectional context, and single-pass decoding, while…
Target Concrete Score Matching: A Holistic Framework for Discrete Diffusion
Ruixiang Zhang, Shuangfei Zhai, Yizhe Zhang +4
Discrete diffusion is a promising framework for modeling and generating discrete data. In this work, we present Target Concrete Score Matching (TCSM), a novel and versatile objecti…
On the Limited Generalization Capability of the Implicit Reward Model Induced by Direct Preference Optimization
Yong Lin, Skyler Seto, Maartje ter Hoeve +6
Reinforcement Learning from Human Feedback (RLHF) is an effective approach for aligning language models to human preferences. Central to RLHF is learning a reward function for scor…
Improving GFlowNets for Text-to-Image Diffusion Alignment
Dinghuai Zhang, Yizhe Zhang, Jiatao Gu +4
Diffusion models have become the de-facto approach for generating visual data, which are trained to match the distribution of the training dataset. In addition, we also want to con…
Stabilizing Transformer Training by Preventing Attention Entropy Collapse
Shuangfei Zhai, Tatiana Likhomanenko, Etai Littwin +5
Training stability is of great importance to Transformers. In this work, we investigate the training dynamics of Transformers by examining the evolution of the attention layers. In…
Improving Disentangled Text Representation Learning with Information-Theoretic Guidance
Pengyu Cheng, Martin Renqiang Min, Dinghan Shen +4
Learning disentangled representations of natural language is essential for many NLP tasks, e.g., conditional text generation, style transfer, personalized dialogue systems, etc. Si…