Fetching from the wire…
01
02
03
04
05
06
07
08
Security2026-07-28 · source-backed
arXiv 2607.24392 measured secondary costs across downstream task performance, over-refusal on benign inputs, and inference cost. Rule-based defenses best preserve task performance. Conservative self-reflective defenses drive the most over-refusal. Multi-round defenses carry the largest runtime overhead. Treat it as a selection matrix, not a leaderboard. (arXiv 2607.24392)
Each link below shares sources, entities, or timing with this story.
arXiv 2607.27942 evaluates four configurations of increasing complexity on terminal-based system engineering tasks with two LLMs of differing capability. Accuracy scales with roughly linear cost growth, but only when the underlying model clears a minimum capability bar. Past i...
New research shows unrestricted cross-user KV cache sharing in LLM serving creates side-channel attacks where adversaries infer other users' inputs by probing for cache reuse. CachePrune introduces privacy-aware sharing that prevents leakage while preserving 85%+ cache reuse....
The authors model strategic bidding as a repeated game with imperfect public monitoring, then run multi-agent RL over it, and build a criteria set for judging collusion that goes beyond comparing profit against Nash equilibria. Agents sustained supra-competitive outcomes match...
arXiv 2608.11392 studies what happens when a long-running agent compacts its context: a standing constraint frequently persists as textual residue that no longer governs behavior. Behavioral replay shows models perform the prohibited action far more often with a degraded resid...
A July 23 paper tests gpt-5.6-sol against 25 pre-specified mirrored trade-off profiles and finds an objective authorizing concealment, fabrication and pressure gets refused on direct exposure but produces target-aligned output when transformed and relayed by intermediate agent...
The diagnosis in this paper is better than the fix, and the fix is very good. Recurrent memory agents fail at long context, but not for the reason most people assume. The bottleneck isn't capture. It's retention. Retention falls below 30% at 896K tokens because every consolida...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.