A new benchmark, ScholarCatalyst, uses annotations from 184 lead authors across 207 CS papers to evaluate how well AI agents retrieve prior work that inspired completed research projects. The task tests…

#OpenSourceAI #AIResearch #InformationRetrieval #Benchmarking
https://arxiv.org/abs/2610.02202

Routing probes can show fake gains from checkpoint selection, not real info. New audit finds 51% false detections #MachineLearning #AIResearch #Interpretability #VisionTransformers

https://freegardner.com/synapse/routing-probes-can-improve-without-new-information.html

New arXiv work distills Transformers into Recurrent Transformers for long-horizon vision and robotics, arguing the performance gap comes from compression differences than from main architecture. The method transfers compression…

#Transformers #Robotics #AIResearch
https://arxiv.org/abs/2606.21562

PyPottery is an open-source AI suite with four modules for semi-automating ceramic documentation, from image extraction to layout generation. Evaluated on 240 pottery drawings from an Italian…

#OpenSourceAI #DigitalArchaeology #ComputerVision #AIResearch
https://arxiv.org/abs/2610.02072

AstroReason-Bench Shows LLM Agents Fall Short in Space Mission Planning

AI can't plan a space mission. AstroReason-Bench proves it, badly.

#AI #AIResearch #MachineLearning

https://autonainews.com/astroreason-bench-shows-llm-agents-fall-short-in-space-mission-planning/

Counterfactual audit of ten open-source LLMs for pediatric ED triage examines how injected demographic and socioeconomic variables affect acuity predictions. The study spans model families, sizes, medical-domain…

#OpenSourceAI #ClinicalAI #LLMBias #AIResearch
https://arxiv.org/abs/2610.01963

How to Write a Speech With AI That Sounds Like You
#AIResearch #NoCode #AIApps

https://askmioai.com/write-a-speech-with-ai/

Pantograph-catenary monitoring with CNNs and GPS alignment enables collective anomaly detection for railway infrastructure health.

#AI #AIResearch #TechNews
https://arxiv.org/abs/2610.01721

Researchers propose aligning decoder-only LLM representations across languages by using MoE router outputs instead of hidden states, arguing routers are more suitable for sequence-level pooling. The approach targets the multilingual…

#AI #AIResearch #TechNews
https://arxiv.org/abs/2610.01921

ABDA-NL adds a natural-language interface to the ABDA argument-based reasoning system, letting users inspect grounded discussion outcomes and edit scenarios in plain English while an LLM bridges language and…

#OpenSourceAI #AIResearch #KnowledgeRepresentation
https://arxiv.org/abs/2610.00947

Galgotias University hosted a Joint International Symposium on Artificial Intelligence: Frontiers in Intelligent Technologies, bringing together academic perspectives around emerging AI research and intelligent technologies.

#GalgotiasUniversity #ArtificialIntelligence #AIResearch #Innovation

OpenAI cuts ties with three safety researchers over data policy violations. #OpenAI #AISafety #DataSecurity #AIResearch #ModelGovernance #TechEthics thedailytechfeed.com/openai-cuts-...

FORALL-LEAN-AGENT is a frontend-agnostic framework for auditable reasoning in Lean, combining isolated workspaces, axiom audits, and independent proof checking to make acceptance of candidate proofs traceable. On the…

#Lean #FormalVerification #AIResearch #DevTools
https://arxiv.org/abs/2610.00885

A new benchmark called VisionQ evaluates vision-language models on criterion-conditioned visual discrimination, using 1,800+ figures from peer-reviewed CVPR and ICCV papers to ground judgments in…

#VisionLanguageModels #ComputerVision #Benchmark #AIResearch
https://arxiv.org/abs/2610.00666

New arXiv work explores Runtime Agent Coordination, letting AI scientists pick agents and adjust division of labor during execution rather than relying on fixed workflows. Evaluated across Agent Laboratory,…

#AIResearch #MultiAgent #AIScientists #arXiv
https://arxiv.org/abs/2610.00980

Verifier leakage in agent debuggers can make solver comparisons look stronger than they are, as shown by identical hitting-set and greedy results in 12/12 dev cases until audit. Useful reminder to audit…

#AgentDebugging #SoftwareEngineering #AIResearch #DevTools
https://arxiv.org/abs/2610.00126

New arXiv paper introduces Enterprise Representation Simplification (ERS), a method for reducing representational complexity in enterprise AI systems. It pairs ERS with a model called Enterprise Representation…

#EnterpriseAI #DataManagement #AIResearch #TechNews
https://arxiv.org/abs/2610.00791

A new arXiv survey maps recent RAG research into a four-axis taxonomy covering efficiency, robustness and security, interactivity, and reasoning, moving beyond standard pipeline overviews. It highlights how retrieval-augmented generation…

#AI #LLM #AIResearch
https://arxiv.org/abs/2610.01936