Tracked topic
…while GPT-5.4 achieves 51.90% accuracy in block-free settings, it collapses to 11.36% under the most severe blocking condition. Further analysis shows that agents are especially vulnerable when…
…Recent preference alignment methods typically rely on supervision distilled from stronger models such as GPT. However, this offline paradigm introduces a Supervision-Perception Mismatch: the student model is forced to align with…
…Our trained Qwen3-VL-8B model improves over Qwen-3-VL-8B-Thinking from 2.8\% to 33.3\% tree-generation accuracy on CurveBench-Easy, exceeding GPT-5.4 and Claude Opus…
…With Qwen3-4B as the backbone, our framework achieves the strongest aggregate performance on our benchmarks, outperforming larger proprietary LLMs (e.g., GPT, Gemini) and fixed-environment training baselines. We further analyze…
…Experiments show that MemPrivacy achieves strong performance in privacy information extraction , substantially surpassing strong general-purpose models such as GPT-5.2 and Gemini-3.1-Pro, while also reducing inference latency…
…Trained from scratch on OpenWebText2 , TextLDM substantially outperforms prior diffusion language models and matches GPT-2 under the same settings. Our results establish that the visual DiT recipe transfers effectively to language…
…using only around 1k samples, achieving comparable performance with leading closed-source models (e.g., GPT-5.4). Based on AgentDoG 1.5 , we construct a highly efficient agentic safety SFT and…
…Using only 0.05% of the EditReward preference data, RewardHarness achieves 47.4% average accuracy on image-editing evaluation benchmarks, surpassing GPT-5 by 5.3 points. When used as a reward…
…Extensive evaluations on 20 LLMs across 7 families (including frontier models such as GPT-5.4, Gemini-3.1-Pro, and Claude-Opus-4.7) demonstrate unsatisfactory performance, with scores below 60…
…On SCOUT-450 , a safety-oriented operating point reduces attack-success rate by 46% and total wall-clock by 40% relative to an always-on GPT-4o judge , at a 5.1…