Vending-Bench: A Benchmark for Long-Term Coherence of Autonomous Agents (arxiv.org)

Screenshot of arxiv.org: Vending-Bench: A Benchmark for Long-Term Coherence of Autonomous Agents

First screen, captured 1 year ago.

Looks similar

Screenshot of arxiv.org: S-LoRA: Serving Concurrent LoRA Adapters

S-LoRA: Serving Concurrent LoRA Adapters (arxiv.org)

28 points by Palmik 3 years ago | 4 comments

Screenshot of arxiv.org: Frontier AI agents violate ethical constraints 30–50% of time, pressured by KPIs

Frontier AI agents violate ethical constraints 30–50% of time, pressured by KPIs (arxiv.org)

512 points by tiny-automates 8 months ago | 329 comments

arXiv paper introducing a 40-scenario benchmark measuring how autonomous AI agents violate ethical, legal, or safety constraints when pressured by KPI-driven outcomes.

ai-agents ai-safety benchmarks alignment

Screenshot of arxiv.org: MemGPT: Towards LLMs as Operating Systems

MemGPT: Towards LLMs as Operating Systems (arxiv.org)

203 points by belter 3 years ago | 105 comments

Screenshot of arxiv.org: LLM agents can autonomously hack websites

LLM agents can autonomously hack websites (arxiv.org)

56 points by pella 2 years ago | 16 comments

Screenshot of arxiv.org: Handbook.md shows that long policy documents do not reliably govern agents

Handbook.md shows that long policy documents do not reliably govern agents (arxiv.org)

309 points by spIrr 2 months ago | 186 comments

Screenshot of arxiv.org: AutoHete: An Automatic and Efficient Heterogeneous Training System for LLMs

AutoHete: An Automatic and Efficient Heterogeneous Training System for LLMs (arxiv.org)

41 points by PaulHoule 1 year ago | 0 comments

Looks similar

S-LoRA: Serving Concurrent LoRA Adapters(arxiv.org)

Frontier AI agents violate ethical constraints 30–50% of time, pressured by KPIs(arxiv.org)

MemGPT: Towards LLMs as Operating Systems(arxiv.org)

LLM agents can autonomously hack websites(arxiv.org)

Handbook.md shows that long policy documents do not reliably govern agents(arxiv.org)

AutoHete: An Automatic and Efficient Heterogeneous Training System for LLMs(arxiv.org)