Tracked topic
…Our evaluation of nine configurations across five model families -- GPT-5.4, Claude Sonnet 4.6, Gemini 2.5 Flash, DeepSeek V4 Flash, Qwen3-32B and GPT-5 mini -- reveals a steep…
…In this paper, we moved beyond static, single-turn prompts to analyze multi-turn adversarial dialogues across distilled models like DeepSeek-R1-7B, Phi-4-Mini, and Qwen-4B-Thinking. By introducing…
…It also generalizes across models and harnesses, improving DeepSeek-V4-Flash (OpenCode) from 12.13 to 19.58. View arXiv page View PDF GitHub 14 Add to collection Community AutoTrainess: Teaching Language…
…According to DeepSeek's paper, DeepSeek-Distill-Qwen-7B's performance in MATH-500 and AIME24 is 92.8 and 55.5 respectively, which seems to be very different from the values…
…Compared with 1T-parameter model such as Kimi-K2.6 and DeepSeek-V4-pro, Agents-A1 achieves leading results on SEAL-0 (56.4), IFBench (80.6), HiPhO (46.4), FrontierScience-Olympiad…
…I tested out the new DeepSeek-R1-Distill-Llama-70B-Uncensored-v2-Unbiased model yesterday. It was a very crude test, but I was quite impressed. I'm a newb over here…
…We evaluate across two forget categories, biohazardous knowledge and abusive tendencies , and four model families spanning dense and Mixture-of-Experts architectures (Llama 3, Qwen 3.5, Gemma 4 E4B, DeepSeek V2…
…We find that the choice of scoring method is the most impactful, with our novel diversity-aware scoring consistently outperforming prior methods on Qwen3-30B-A3B, DeepSeek-V2-Lite, and GPT-OSS…
…pooled Spearman rho = +0.436, p < 0.001 on DeepSeek+Codestral; rho = +0.593 on Codestral alone; rho = +0.356 on Qwen. This establishes Delta H as a cheap, rewriter-agnostic proxy…
…blog post, check it out! This podcast is generated via ngxson/kokoro-podcast-generator , using DeepSeek-R1 and Kokoro-TTS can Math-Verify be integrated into lighteval library ? Hi @ pulkitmehtawork ! It already…