Technical Report: Performance and baseline evaluations of gpt-oss-safeguard-120b and gpt-oss-safeguard-20b
gpt-oss-safeguard-120b and gpt-oss-safeguard-20b are two open-weight reasoning models post-trained from the gpt-oss models and trained…
Author: AI News Team
Introducing gpt-oss-safeguard
OpenAI introduces gpt-oss-safeguard—open-weight reasoning models for safety classification that let developers apply and iterate on…
RL for Reasoning by Adaptively Revealing Rationales
We propose that reinforcement learning (RL) from partial expert demonstrations is not merely a training…
Reasoning’s Razor: Reasoning Improves Accuracy but Can Hurt Recall at Critical Operating Points in Safety and Hallucination Detection
Reasoning has become a central paradigm for large language models (LLMs), consistently boosting accuracy across…
Toward Machine Interpreting: Lessons from Human Interpreting Studies
Current speech translation systems, while having achieved impressive accuracies, are rather static in their behavior…
Inside CampusAI’s mission to close the AI training gap for everyday workers — check it out at TechCrunch Disrupt 2025
CampusAI is like “Coursera meets GitHub meets Tinder in the metaverse” — an educational platform…
Mappa’s AI voice analysis helps you find the best job candidates and will show off its tech at TechCrunch Disrupt 2025
Mappa's AI hiring platform can assess a candidate's behavior based on voice patterns, and aims…
Super Teacher is building an AI tutor for elementary schools — catch it at Disrupt 2025
Super Teacher’s app features animated tutors with AI-generated voices that guide students through interactive lessons.…
Upcoming Workshop on Post-AGI Economics, Culture, and Governance
Published on October 28, 2025 9:55 PM GMTThis is an announcement and call for applications…