Generalize across Homophily and Heterophily: Hybrid Spectral Graph Pre-Training and Prompt Tuning

[ad_1] arXiv:2508.11328v1 Announce Type: cross Abstract: Graph “pre-training and prompt-tuning” aligns downstream tasks with pre-trained objectives to enable efficient knowledge ...
Read more
ORFuzz: Fuzzing the "Other Side" of LLM Safety — Testing Over-Refusal

[ad_1] arXiv:2508.11222v1 Announce Type: cross Abstract: Large Language Models (LLMs) increasingly exhibit over-refusal – erroneously rejecting benign queries due to ...
Read more
Optimizing Rotation Transformation for Quantization for Free

[ad_1] [Submitted on 2 May 2025 (v1), last revised 14 Aug 2025 (this version, v2)] View a PDF of the ...
Read more
Unlocking Adaptive Serial-Parallel Decoding by Exploring Intrinsic Parallelism in LLMs

[ad_1] [Submitted on 12 Aug 2025 (v1), last revised 14 Aug 2025 (this version, v2)] View a PDF of the ...
Read more
AI Blob! LLM-Driven Recontextualization of Italian Television Archives

[ad_1] arXiv:2508.09535v1 Announce Type: cross Abstract: This paper introduces AI Blob!, an experimental system designed to explore the potential of ...
Read more
[2508.06671] Do Biased Models Have Biased Thoughts?

[ad_1] [Submitted on 8 Aug 2025 (v1), last revised 12 Aug 2025 (this version, v2)] View a PDF of the ...
Read more
A Framework for Deep Research Agents and Agent Foundation Models Training

[ad_1] [Submitted on 1 Aug 2025 (v1), last revised 12 Aug 2025 (this version, v2)] Authors:Tianqing Fang, Zhisong Zhang, Xiaoyang ...
Read more
Generative AI for Strategic Plan Development

[ad_1] arXiv:2508.07405v1 Announce Type: cross Abstract: Given recent breakthroughs in Generative Artificial Intelligence (GAI) and Large Language Models (LLMs), more ...
Read more
Uncovering the Emerging Prosody Processing Capabilities in Speech Language Models

[ad_1] [Submitted on 27 Jul 2025 (v1), last revised 7 Aug 2025 (this version, v2)] View a PDF of the ...
Read more
A Large-scale Instruction Tuning Dataset for Code LLMs

[ad_1] [Submitted on 5 Apr 2025 (v1), last revised 7 Aug 2025 (this version, v2)] View a PDF of the ...
Read more









