Tuesday, June 16: 7 frontier papers out of 658 read.
Papers rated as moving the field
How we choose“Nemotron 3 Ultra achieves up to 6x higher inference throughput as compared to state-of-the-art publicly available LLMs while attaining on-par accuracy.”
The authors, in the abstract
Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
“We show that reinforcement learning can make a policy \emph{addicted} to such a visible self-benefit channel.”
“Evaluating 15 advanced agents reveals that robust spatial task solving remains challenging: the strongest model, GPT-5, achieves an average task success rate (TSR) of only 17.4%, while the leading open-source model, Qwen-3.5, reaches 14.1%.”
“Our results show that reward hacking arises naturally when optimizing proxy objectives with capable language model agents and resists standard mitigations, suggesting that proxy-reward failures in agentic settings may require approaches beyond standard exploration and credit-assignment fixes.”
“Extensive experiments demonstrate that our method achieves surgical concept erasure performance across various domains with minimal training overhead, timely and elegently resolute the critical safety vulnerabilities inherent in emerging VAR architectures.”
Jun 10–16: 208 cleared, 15 frontier
The week’s papers have not been sorted into fields yet.
Papers this week that name the model
Papers this week with an author at the lab
57 papers