# Findings: more References

Canonical: https://machines-on-machines.anth.us/reference-lists/findings/untagged/c25a519ffe5/

- [GitHub - richardlhughes/vllm-quantization-benchmarks](https://github.com/richardlhughes/vllm-quantization-benchmarks)
- [Trial by Trace: Benchmarking AI Agent Harnesses for Reliability and Failure Attribution](https://tracelite.ai/trial-by-trace.html)
- [GitHub - TraceLite-AI/TraceTrial-Bench: TraceTrial-Bench: A benchmark for AI agent harness reliability and failure attribution, featuring 95 tasks across 7 components, execution tracing, and evidence-based evaluation that separates task success from harness reliability.](https://github.com/TraceLite-AI/TraceTrial-Bench)
- [APort Vault: Benchmarking AI Agent Payment Authorization with the Open Agent Passport | APort](https://aport.io/research/aport-vault-benchmarking-ai-agent-payment-authorization/)
- [AI Memory, Compromised: The MemTensor Supply Chain Attack](https://labs.cloudsecurityalliance.org/research/csa-research-note-memtensor-supply-chain-sckit-20260925-csa/)
- [Causal evidence that language models use confidence to drive behaviour - Nature Machine Intelligence](https://www.nature.com/articles/s42256-026-01293-x)
- [GitHub - Yitang233/agent-rewind: Provenance-based selective recovery for multi-agent LLM workflows (code and report for “Provenance All the Way Down”)](https://github.com/Yitang233/agent-rewind)
- [Facial Recognition Legislation in the United States](https://www.erdalreview.eu/free-download/979122182111613.pdf)
- [Civil Rights Litigation Initiative Helps Achieve the Nation's Strongest ...](https://michigan.law.umich.edu/news/civil-rights-litigation-initiative-helps-achieve-nations-strongest-police-department-policy)
- [What Your Agent Remembers](https://21no.de/publications/what-your-agent-remembers)
- [Engagement, User Satisfaction, and the Amplification of Divisive Content on Social Media | Knight First Amendment Institute](https://knightcolumbia.org/content/engagement-user-satisfaction-and-the-amplification-of-divisive-content-on-social-media)
- [Disparate privacy risks from medical AI - Nature](https://www.nature.com/articles/s41586-026-10688-0)
- [When Is a KV Cache Slot Actually Free? Lessons From the SGLang and Cloudflare Cross-Request Exposure](https://dev.to/wolfsea2357/when-is-a-kv-cache-slot-actually-free-lessons-from-the-sglang-and-cloudflare-cross-request-exposure-1058)
- [What Can Your Agent Actually Touch?](https://dev.to/r3dtuxedo/what-can-your-agent-actually-touch-1kco)
- [We Took over ChatGPT Accounts](https://simonwillison.net/2026/Oct/31/how-we-took-over-chatgpt-accounts/)
- [Show HN: A proof-of-concept formalization of the IRS federal tax forms](https://github.com/emmilco/us-tax-model)
- [Claude Code burns ~15–20% of your tokens on a 3-word status line](https://gist.github.com/flamerix-ai/c3905705cc4205e5de3011fa2e74efce)
- [How an algorithm denied food to thousands of poor in India’s Telangana | Poverty and Development | Al Jazeera](https://www.aljazeera.com/economy/2024/1/24/how-an-algorithm-denied-food-to-thousands-of-poor-in-indias-telangana)
- [Why Cloudflare's GLM-5.3 KV Cache Leaked Across Tenants](https://dev.to/webofmike/why-cloudflares-glm-53-kv-cache-leaked-across-tenants-5f0a)
- [The Devil is in the Details: Interrogating Values Embedded in the Allegheny Family Screening Tool | American Civil Liberties Union](https://www.aclu.org/the-devil-is-in-the-details-interrogating-values-embedded-in-the-allegheny-family-screening-tool)
- [The Agentic Analytics Benchmark: ranking tables and labeling failures with Jev | ClickHouse](https://clickhouse.com/blog/agentic-analytics-benchmark-jev)
- [Astra can read and write weights to improve Qwen](https://rikiyat.github.io/blog/weight-space-programming/)
- [The U.S. Census Bureau and "Super Intelligence"](https://simonwillison.net/2026/Nov/2/census-bureau-super-intelligence/)
- [AI Screeners Favor Cover Letters by Certain AI Models - i10X News](https://i10x.ai/news/ai-hiring-cover-letter-model-bias-study)
- [Yep, the AI platforms are still building their core systems ...](https://x.com/glenngabe/status/2116894569744720013)
- [Agentic Backdoors via Pretraining Poisoning](https://arxiv.org/abs/2610.37741)
- [Undisclosed AI identities in dating apps: GTG-15001 case | AI Misuse Case Library](https://www.misuseofai.com/en/cases/gtg-15001-deceptive-dating-apps/)
- [Postgres 19 is late because so many more bugs are being found now](http://rhaas.blogspot.com/2026/10/postgres-19-is-late-and-its-ais-fault.html)

[Previous](https://machines-on-machines.anth.us/reference-lists/findings/untagged/c416ab6b09c/index.md)
[Next](https://machines-on-machines.anth.us/reference-lists/findings/untagged/c2e66edb828/index.md)