ML Newsnew | past | best | rss | submit about
Stories from September 1, 2026 (UTC)
Go back a day, month, or year. Go forward a day.
1. Puro-2B: Poor Lab's Qwen2-1.5B Trained on RTX 5090 within $5090 (arxiv.org)
3 points by simonpure 32 days ago | hide | past | pdf | discuss
2. Can escalation channels redirect reward hacking toward defect disclosure? (arxiv.org)
2 points by sbulaev 32 days ago | hide | past | pdf | discuss
3. Program Learning with Verifiable Rewards: Symbolic Backpropagation (arxiv.org)
2 points by sbulaev 32 days ago | hide | past | pdf | discuss
4. Sliding-window beats linear attention (arxiv.org)
2 points by simonpure 32 days ago | hide | past | pdf | 1 comment
5. Logic and the 2-Simplicial Transformer (2019) (arxiv.org)
2 points by wslh 32 days ago | hide | past | pdf | discuss
6. SKILL.state: Scalable Long-Horizon Agent Skills (arxiv.org)
1 point by Hoefner 32 days ago | hide | past | pdf | discuss
7. Ghost Couple:Correlated LLM Name Priors Haunting the Web and Academic Publishing (arxiv.org)
1 point by bookofjoe 32 days ago | hide | past | pdf | discuss
8. Temporal Analysis of NetFlow Datasets for Network Intrusion Detection Systems (arxiv.org)
1 point by ENOMEM 32 days ago | hide | past | pdf | discuss
9. Breaking Execution Continuity of Agent Systems via Rollback (arxiv.org)
1 point by sbulaev 32 days ago | hide | past | pdf | discuss
About | RSS | RSS (all) | HN arXiv