June 19–25: 9 frontier papers out of 2,056 read.
Papers rated as moving the field
How we choose“In the one-million-token context setting, DeepSeek-V4-Pro requires only 27% of single-token inference FLOPs and 10% of KV cache compared with DeepSeek-V3.2.”
The authors, in the abstract
Towards Highly Efficient Million-Token Context Intelligence
“Experiments show that current LLMs still struggle on HOLMES, with an average accuracy of only 50.64% and the best model reaching 59.54%.”
“Evaluated on 457 AdvBench prompts across four GPT models, OTTER raises average ASR from 7.0% to 84.0%.”
“Comparative case studies of two open-source frameworks demonstrate that evolution-native design activates 3.5× more attack surface cells and achieves a 100% attack persistence rate (40/40 payloads across all CIA+Privacy categories), while co-located security scanners block only 2.5% of attacks.”
“Cross-architecture transfer via ridge regression maps yields large behavioral suppression (up to 46 points) but fails specificity controls as random and orthogonal directions perform comparably.”
“Experiments across multiple LALMs demonstrate that effective auditory scene understanding requires integration over all auditory layers, rather than reliance on speech or sound alone, underscoring the necessity of CASU for advancing complex audio understanding in LALMs.”
“Empirical results demonstrate that Qwen-AgentWorld significantly outperforms existing frontier models.”
Jun 19–25: 169 cleared, 9 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
160 papers