From Knowledge Access to Source Learning: Developing Source-Specific Competence

Large language model (LLM) agents increasingly rely on persistent external sources to solve sequences of knowledge-intensive tasks. Existing methods improve how source content is accessed and organized, while agent-memory systems preserve reusable knowledge from prior interactions, but repeated use of the same source is still largely treated as repeated access rather than an opportunity to…

arxiv.org ·2026-10-01

We haven't written up this one. arXiv cs.AI has the full story — the link below goes straight to it.

Read the original at arxiv.org →

More in AI

Finetuning with Sampling: SFT Learns Better Than You Think

Introducing new capabilities to frontier models has long been the goal of posttraining, which predominantly employs supervised finetuning (SFT) and reinforcement learning (RL) to this end.

2026-10-01 ·arxiv.org

MIRTO: a registration-gated, multiverse-tested evaluation protocol for unsupervised anomaly segmentation in brain MRI

Unsupervised anomaly detection (UAD) methods for brain MRI are ranked by a single score, yet that score rests on choices that are rarely reported: how each anomaly map is aligned with the reference…

2026-10-01 ·arxiv.org

Local Support Learning

We explore catastrophic forgetting in the context of large pre-trained models. By considering forgetting as a geometric problem in the input space of each weight matrix, we uncover a natural retention…

2026-10-01 ·arxiv.org

Opus 5.5 loves to tell you ‘this matters’ (and other AI writing tells)

Opus 5.5’s biggest tell is the word “dependable,” which pops up 23 times more often than in human samples.

"Dependable" appears 23 times more often in Claude Opus 5.5's writing than in humans
"This matters" occurs 116 times more frequently in Opus 5.5 than in human writing
Astra model emphasizes "another dimension" and uses "may provide" or "can provide" hedging

2026-10-01 ·techcrunch.com

It’s Looking Possible That Zuckerberg Is Accidentally Committing Corporate Suicide

"The largest misallocation of capital in history." The post It’s Looking Possible That Zuckerberg Is Accidentally Committing Corporate Suicide appeared first on Futurism .

2026-10-01 ·futurism.com

DuoMind: Enabling Distributed Multi-Robot Coordination with Semantic Communication

Vision-language models (VLMs) and vision-language-action models (VLAs) have recently driven rapid progress in general-purpose robots, yet most progress has focused on single-robot settings.

2026-10-01 ·arxiv.org