9 papers
Learning Rules, Memorizing Facts: An Information-Theoretic View of Hallucination
Gengze Xu, Anxin Guo, Yifei Hu +3
Understanding the relationship between generalization and hallucination requires clarifying the role of memorization. We study this relationship from an information-theoretic persp…
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression
DeepSeek-AI, :, Anyi Xu +585
The widespread adoption of long-horizon agents has made model workloads increasingly input-heavy. Although prior work has substantially reduced the cost of long-context computation…
Beyond Penalization: Diffusion-based Out-of-Distribution Detection and Selective Regularization in Offline Reinforcement Learning
Qingjun Wang, Hongtu Zhou, Hang Yu +5
Offline reinforcement learning (RL) faces a critical challenge of overestimating the value of out-of-distribution (OOD) actions. Existing methods mitigate this issue by penalizing…
Neural-Driven Image Editing
Pengfei Zhou, Jie Xia, Xiaopeng Peng +15
Traditional image editing typically relies on manual prompting, making it labor-intensive and inaccessible to individuals with limited motor control or language abilities. Leveragi…
On Weak-to-Strong Generalization and f-Divergence
Wei Yao, Gengze Xu, Huayi Tang +4
Weak-to-strong generalization (W2SG) has emerged as a promising paradigm for stimulating the capabilities of strong pre-trained models by leveraging supervision from weaker supervi…
Weak-to-Strong Generalization via Bregman Bias-Variance Decomposition
Gengze Xu, Wei Yao, Ziqiao Wang +1
Weak-to-strong generalization (W2SG) is the phenomenon in which a powerful student model, trained on labels produced by a weaker teacher, ultimately outperforms the teacher on the…
DD-Ranking: Rethinking the Evaluation of Dataset Distillation
Zekai Li, Xinhao Zhong, Samir Khaki +49
In recent years, dataset distillation has provided a reliable solution for data compression, where models trained on the resulting smaller synthetic datasets achieve performance co…
Ensemble Debiasing Across Class and Sample Levels for Fairer Prompting Accuracy
Ruixi Lin, Ziqiao Wang, Yang You
Language models are strong few-shot learners and achieve good overall accuracy in text classification tasks, masking the fact that their results suffer from great class accuracy im…
MutualVPR: A Mutual Learning Framework for Resolving Supervision Inconsistencies via Adaptive Clustering
Qiwen Gu, Xufei Wang, Junqiao Zhao +4
Visual Place Recognition (VPR) enables robust localization through image retrieval based on learned descriptors. However, drastic appearance variations of images at the same place…