Aamer Mihaysi

O96a's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

commented a paper about 3 hours ago

LoopArena: Benchmarking Models as Runtime Controllers for Loop Engineering

Paper • 2608.28281 • Published 4 days ago • 79 •

2

commented a paper 1 day ago

What Does an Evaluation License? A Commit-Bound Census of Claim-Relative Inference in Inspect Evals

Paper • 2608.19269 • Published 7 days ago • 3 •

3

commented a paper 2 days ago

Agentic Game Development as a Verifiable Trajectory Data Engine for Scaling World Models

Paper • 2608.25518 • Published 6 days ago • 185 •

4

commented a paper 3 days ago

PILOT in the Loop: Live Self-Improvement for Long-Horizon Agents

Paper • 2608.26530 • Published 5 days ago • 30 •

3

commented a paper 5 days ago

TorchMorph: CUDA-accelerated Morphological Transforms

Paper • 2608.24738 • Published 7 days ago • 4 •

3

commented a paper 6 days ago

Quantization-Aware Healing: A Practical Recipe for Recovering Compressed, 4-Bit LLMs

Paper • 2608.20953 • Published 11 days ago • 11 •

4

commented a paper 7 days ago

Daedalus-150M: A Convolution-Attention Hybrid Designed for CPU Inference

Paper • 2608.20210 • Published 12 days ago • 8 •

4

commented a paper 8 days ago

QuoteBench: How Matched Scores Can Hide Command-Path Failures

Paper • 2608.13547 • Published 19 days ago • 8 •

4

commented a paper 9 days ago

The Embedder's Dilemma: LLMs Are Better, but at What Cost?

Paper • 2608.12875 • Published 19 days ago • 15 •

5

commented a paper 10 days ago

SWE-bench Science: Can Coding Agents Resolve Engineering Tasks in Science?

Paper • 2608.19799 • Published 12 days ago • 65 •

4

commented a paper 11 days ago

SkillGate: Training In-Policy Skill Selection in Long-Horizon Agents

Paper • 2608.18852 • Published 13 days ago • 7 •

3

commented a paper 13 days ago

How Do Agents Fail on AutoResearch: End-to-End Diagnostic Evaluation on 100 Real-World Frontier Research Tasks

Paper • 2608.14905 • Published 18 days ago • 30 •

4

commented a paper 14 days ago

Second Thought: Reasoning in Parallel as LLM Agents Act and Observe

Paper • 2608.13667 • Published 18 days ago • 16 •

4

commented a paper 17 days ago

Specification-first convergence with an AI coding agent: a case study of dismantling a core architectural invariant across 189 files in a 717k-line codebase with no test oracle and no human code review

Paper • 2608.12440 • Published 20 days ago • 10 •

5

commented a paper 18 days ago

Ready Cohorts: Bounding GPU Opportunity and Avoiding Host Round Trips in LLM-Agent Control

Paper • 2608.12123 • Published 20 days ago • 2 •

3

commented a paper 19 days ago

Not Worth Another Token: Marginal Value Estimation for Efficient Deep Research Agents

Paper • 2608.08389 • Published 23 days ago • 12 •

3

commented 3 papers 20 days ago

Business Arena: Benchmarking LLM Agents in a Realistic Marketplace

Paper • 2608.08621 • Published 23 days ago • 22 •

5

Omega-S: A Functional Resilience Index for LLM Fine-Tuning

Paper • 2608.03887 • Published 28 days ago • 7 •

5

Stealing Reasoning Traces from Proprietary LLM APIs

Paper • 2608.09867 • Published 22 days ago • 117 •

3

commented a paper 21 days ago

Relevant but Incomplete: Referential Dangling as a Paradigm-Level Failure Mode in Hard Prompt Compression

Paper • 2608.04569 • Published 27 days ago • 14 •

3