11 papers
The Physics Is the Network: Fast, Training-Free Neural Emulation
Luka Todorovic, Kaushal Gianchandani, Sai Ravela
Neural emulation need not require learning a substitute for known physics. We investigate PolyNets that couple a continuous-time neural representation of polynomial dynamics with a…
Multi-Rollout On-Policy Distillation via Peer Successes and Failures
Weichen Yu, Xiaomin Li, Yizhou Zhao +8
Large language models are often post-trained with sparse verifier rewards, which indicate whether a sampled trajectory succeeds but provide limited guidance about where reasoning s…
UniSD: Towards a Unified Self-Distillation Framework for Large Language Models
Yiqiao Jin, Yiyang Wang, Lucheng Fu +7
Self-distillation (SD) offers a promising path for adapting large language models (LLMs) without relying on stronger external teachers. However, SD in autoregressive LLMs remains c…
TorchUMM: A Unified Multimodal Model Codebase for Evaluation, Analysis, and Post-training
Yinyi Luo, Wenwen Wang, Hayes Bai +6
Recent advances in unified multimodal models (UMMs) have led to a proliferation of architectures capable of understanding, generating, and editing across visual and textual modalit…
LatentUMM: Dual Latent Alignment for Unified Multimodal Models
Yinyi Luo, Wenwen Wang, Hayes Bai +2
Unified multimodal models (UMMs) achieve strong performance in both understanding and generation by learning a shared latent space, yet they often exhibit functional inconsistency…
UniPath: Adaptive Coordination of Understanding and Generation for Unified Multimodal Reasoning
Hayes Bai, Yinyi Luo, Wenwen Wang +2
Unified multimodal models (UMMs) aim to integrate understanding and generation within a single architecture. However, it remains underexplored how to effectively coordinate these t…
Self-Corrected Image Generation with Explainable Latent Rewards
Yinyi Luo, Hrishikesh Gokhale, Marios Savvides +2
Despite significant progress in text-to-image generation, aligning outputs with complex prompts remains challenging, particularly for fine-grained semantics and spatial relations.…
KnowledgeSmith: Uncovering Knowledge Updating in LLMs with Model Editing and Unlearning
Yinyi Luo, Zhexian Zhou, Hao Chen +4
Knowledge editing and machine unlearning are two popular approaches for large language models (LLMs) to stay up-to-date. However, the knowledge updating mechanism of LLMs remains l…
Image Tokenizer Needs Post-Training
Kai Qiu, Xiang Li, Hao Chen +7
Recent image generative models typically capture the image distribution in a pre-constructed latent space, relying on a frozen image tokenizer. However, there exists a significant…
Robust Latent Matters: Boosting Image Generation with Sampling Error Synthesis
Kai Qiu, Xiang Li, Jason Kuen +7
Recent image generation schemes typically capture image distribution in a pre-constructed latent space relying on a frozen image tokenizer. Though the performance of tokenizer play…
SOLAR: Scalable Optimization of Large-scale Architecture for Reasoning
Chen Li, Yinyi Luo, Anudeep Bolimera +4
Large Language Models excel in reasoning yet often rely on Chain-of-Thought prompts, limiting performance on tasks demanding more nuanced topological structures. We present SOLAR (…