2026-11-08

Long-term memory enables large language model (LLM) agents to provide continuous personalized assistance by retaining user information across interactions. Existing systems support this process through fact extraction, organization, updating, and retrieval, but largely maintain memories as independent facts. This fact-centric paradigm may overlook how new facts update, constrain, or conflict with earlier ones, while leaving their applicability, role, and priority for the current request implicit

Multimodal large language models (MLLMs) can now process hour-long videos, making extended multi-turn dialogue over a single video a natural mode of use. Sustaining such a dialogue requires more than answering each question in isolation: the model must retrieve and reuse information established earlier in the conversation, both what it has seen and what it has said, an ability we term in-context video memory. Existing video QA benchmarks leave this ability unmeasured: their questions are mutuall

🤖VideoMemBench evaluates multimodal language models on long multi-turn video dialogues requiring retrieval of earlier visual and conversational context, revealing substantial performance gaps.

Group Relative Policy Optimization (GRPO) has advanced generative model post-training in verifiable domains such as code and mathematics, but evaluating video quality, especially emotional expression, often depends on subjective aesthetics and has no uniquely correct target. Existing approaches typically rely on fixed rewards for visual quality, motion quality, and text alignment that poorly capture emotional intent and can invite reward hacking, while RLHF is difficult to scale and subject to i

Multi-teacher on-policy distillation (MOPD) extends on-policy distillation (OPD) to consolidate capabilities from multiple teachers. Although prior work reports near-perfect in-distribution (IND) distillation performance, we find that this can mask incomplete capability consolidation beyond the supervised distribution. A preliminary study on synthetic reasoning shows that OPD does not fully transfer teachers' generalization capabilities. We then extend this investigation to real-world tasks, eva

Self-distillation methods for large language models with an exponential moving average (EMA) teacher already retain a history of student updates. Muon nevertheless stores a separate momentum buffer, increasing persistent training memory. We propose Muon with Reused Teacher (Muon-RT), which reuses the teacher-student displacement to eliminate momentum buffers for hidden weight matrices. We show that this displacement encodes an exponentially weighted history of applied updates and smooths tempora

🤖Muon-RT eliminates separate momentum buffers by reusing teacher-student displacement for hidden weight matrices, reducing memory while improving fine-tuning and policy optimization accuracy.

Both AI tools offer to delve deep into your bank accounts, but they differ in privacy settings.

Twenty-five years ago, Kevin Lewandowski founded Discogs, arguably the destination online for obsessive vinyl collectors and music nerds of all stripes. It is a marketplace for buying and selling records, obviously, but it's also become so much more. It's a shockingly comprehensive database that covers not just albums and singles, but all the various versions […]

Article URL: Comments URL: Points: 57 # Comments: 12

Samsung appears to be preparing the One UI 9.5 beta program ahead of an expected 2027 launch. Here's what to do if you want to test Samsung's latest OS.

Developer: “I can’t be part of the Fox world.”

I built Coldframe, a honeypot that pretends to be a supermarket refrigeration controller. Inspired by the events on industrial fridge installations in the past weeks, I wanted to see what people would try against a convincing fake. To learn more about this I built a honeypot that resembled perfectly the Danfoss AK-SM 800A and Dixell XWEB interfaces. Both of them are recognized on Shodan as real ones (an not as honeypots). It’s one Python file using only the standard library, and it builds on epp