AITOPAI热报

#GPT-OSS-20B

4 stories · 1 in 7 days · 3 in 30 days
AI安全 2GRPO 1Llama-3.3-70B-Instruct 1prompt injection 1Qwen2.5 1Qwen3-8B 1SageMaker HyperPod 1价值移植 1
Stories, products and papers tagged "GPT-OSS-20B", newest first.
Nov 10
PapersBuilders10:58
Chinese content论文揭示 LLM 智能体「价值-策略解耦」：内部算出最优动作却不采用

这篇论文做了一件很细的活：拿 Qwen2.5 和 GPT-OSS 在 Grid 和 WebShop 上证明模型内部其实能算出哪个动作更优，但策略层面偏偏不用。对做智能体可靠性或可解释性研究的人来说是很有用的诊断框架，NVT 这种干预思路也值得上手复现。

#Qwen2.5#GPT-OSS-20B#智能体#可解释性
Event
arXiv cs.AI@Yuzhang Luo et al. (4)Source
Oct 28
Oct 21
How-toBuildersOfficialSelected01:52
GPT-OSS-20B fine-tuned for multilingual reasoning on SageMaker HyperPod

The real lesson is the decision framework: pick SFT when accuracy targets are clear, and add GRPO only when benchmarks show remaining headroom, avoiding wasted RL compute. Practitioners building multilingual models on their own data can replicate the workflow on SageMaker HyperPod with open weights, no dependency on a closed API.

#GPT-OSS-20B#Amazon SageMaker#fine-tuning#reasoning models
Primary sourceAWS Machine Learning Blog@Mijanur PalashSource
Sep 29