Knowledge catalogue
arXiv:2610.11132v1 Announce Type: cross Abstract: Supervised fine-tuning (SFT) equips large language models (LLMs) with specialized capabilities, but often comes at the cost of forgetting the general
arXiv:2610.11371v1 Announce Type: new Abstract: Contemporary decoder-only large language models (LLMs) have demonstrated strong capabilities across a wide range of domains. However, existing pretraini
arXiv:2610.10812v1 Announce Type: new Abstract: Small language models (SLMs) have been increasingly adopted for onboard robot operation because they enable intelligent decision-making. However, existi
SlopSoup TV is a live, never-ending pixel-art TV network. Every script, character, voice, camera cut and schedule decision is made by models. It's made to be very absurd, dark and strange. Hardware: o
arXiv:2610.10871v1 Announce Type: new Abstract: Large Language Models (LLMs) achieve strong performance across many domains, but their efficiency is limited by the quadratic cost of attention with res
arXiv:2610.11730v1 Announce Type: new Abstract: Standard weight decay treats each weight matrix as a vector and ignores its spectral structure. We introduce spectral weight decay, a post-step decouple
arXiv:2603.28744v2 Announce Type: replace Abstract: The linear representation hypothesis states that neural network activations encode high-level concepts as linear mixtures. However, under superposit
arXiv:2610.10549v1 Announce Type: new Abstract: Tool-calling agents have become central to enterprise AI, yet training and evaluating them at scale remains severely constrained due to business and leg
arXiv:2610.11136v1 Announce Type: new Abstract: Presupposing the boundaries of bias is itself a form of bias. We study closed-loop bias governance for Dutch government documents, where a system must d
Figure 1b from our paper Disclosure: I'm one of the authors (Microsoft). The paper, code, dataset are public and ThinkingBox is on Hugging Face OpenEnv as well. Raw evaluation trajectories are not rel
TRL v1.15 is out, and it’s an absolute banger of a release for memory-efficient post-training. The main change: SFT, DPO, KTO, GRPO, RLOO and Distillation now use a fused LM head by default. Instead o
arXiv:2610.11322v1 Announce Type: new Abstract: Geometrically faithful and functional articulated 3D assets are essential for real-to-sim robot manipulation, where policies trained in simulation must
arXiv:2610.11123v1 Announce Type: new Abstract: Most human-agent interaction today remains text-based. Natural language can impose cognitive overload, ambiguity, information chaos, and slow input for
arXiv:2610.11286v1 Announce Type: new Abstract: The reliability and accountability of image generative models (IGMs) are essential for building responsible and trustworthy AI systems. Recent IGMs, suc
arXiv:2610.09588v1 Announce Type: new Abstract: Deploying robots as Complex Adaptive Systems (CAS) in unknown and dynamic environments necessitates a transition from rigid command libraries toward int
arXiv:2610.09764v1 Announce Type: new Abstract: Large Language Models (LLMs) can generate drone programs from natural-language mission descriptions, but syntactically valid programs may still violate
arXiv:2610.08875v1 Announce Type: cross Abstract: Agent Skills package procedural guidance and resources for reuse, but a relevant Skill does not necessarily improve task performance. Existing studies
arXiv:2610.10135v1 Announce Type: cross Abstract: Sparse attention mechanisms estimate attention over n tokens using a small subset of keys. Many existing approaches use maximum inner product search (
arXiv:2505.06592v2 Announce Type: replace Abstract: In this paper, we propose batch augmentation with unimodal fine-tuning for multimodal learning. We start with pre-trained unimodal models. We fine-t
arXiv:2610.09193v1 Announce Type: cross Abstract: MemoryAgentBench's Conflict Resolution split is read as measuring 'selective forgetting'. We execute the benchmark's own rule - the newest statement a
Vultr demonstrated its composable cloud platform at the Agent Arena Hackathon in San Francisco by partnering with Cerebral Valley to build production‑ready enterprise agentic AI systems using Serverle
arXiv:2610.09276v1 Announce Type: cross Abstract: Modern DBMSs expose hundreds of configuration knobs, resulting in a high-dimensional and heterogeneous search space that makes automated tuning costly
arXiv:2610.08862v1 Announce Type: cross Abstract: Household robots must accommodate new user instructions while executing ongoing tasks. Existing agents often regenerate or extensively revise the rema
arXiv:2610.10426v1 Announce Type: new Abstract: Terminal-agent capability depends jointly on model weights and the runtime harness that formats prompts, binds tools, and handles error recovery. Existi
arXiv:2610.09321v1 Announce Type: new Abstract: Speech Language Model (SLM) performance often degrades on dialects due to data scarcity. Conventional text-to-speech (TTS) augmentation struggles to cov
arXiv:2610.09907v1 Announce Type: new Abstract: Vision-Language Models (VLMs) have shown strong performance across a wide range of downstream vision tasks, thanks to the complementary information cont
Coverage period: July 1, 2026 through September 30, 2026 Overview This roundup consolidates selected major AI-related security incidents and exploit disclosures reported during the coverage period. It
arXiv:2505.08854v2 Announce Type: replace Abstract: Generative Artificial Intelligence (GenAI) constitutes a transformative technological wave that reconfigures industries through its unparalleled cap
This table to help anyone looking to build a budget Data Center homelab. I copied the bulk of value based, mid level, decent speed results GPUs and feed it to AI or SI and here are the recommended res
Interesting approach to solving the problem of getting your grokbot to talk to other agents Connecting agents together safely and having them talk to each other or your friends agents used to suck. Po
A week in Beijing and Shanghai with the people building AI in China some quotes from the article: ByteDance/Doubao is the only major lab keeping its flagship closed, because it is the only one that al
arXiv:2610.09398v1 Announce Type: new Abstract: In tropical regions, their agricultural sectors remain highly dependent on manual labor for fruit harvesting. On large-scale farms, this dependency ofte
Classification is one of the oldest problems in machine learning. Unlike modern decoding transformers, most classification models were non-autoregressive by nature. Before Jev, the common approaches w
arXiv:2610.10394v1 Announce Type: new Abstract: Kernels encode the inductive bias of a wide range of machine learning models, yet automated kernel design faces a fundamental dilemma. A fixed grammar o
arXiv:2610.10114v1 Announce Type: new Abstract: The architectural design of Large Language Models (LLMs) is shifting from traditional full-attention-only models to hybrid models, which combine differe
arXiv:2610.09209v1 Announce Type: new Abstract: Substance Use Disorder (SUD) counseling requires patient responses that reflect underlying cognitive states such as beliefs, coping strategies, and read
arXiv:2610.10385v1 Announce Type: new Abstract: Embedding tables are among the largest components of modern language models. Most compression methods fix a coding geometry such as coordinate blocks, l
arXiv:2603.11938v2 Announce Type: replace-cross Abstract: Structured radiology reporting promises faster, more consistent communication than free text, but automation remains difficult as models must
arXiv:2610.09227v1 Announce Type: cross Abstract: Mixed-precision post-training quantization needs a per-module sensitivity signal; for a text embedder the obvious one -- the retrieval quality a modul
arXiv:2509.18597v3 Announce Type: replace Abstract: Large language models (LLMs) can translate natural-language instructions for robotic manipulation into executable code, but ambiguity, noisy generat
arXiv:2610.09665v1 Announce Type: new Abstract: On-policy post-training can improve large language models by learning from their own trajectories, but requires costly rollout generation. We ask whethe
arXiv:2610.08887v1 Announce Type: new Abstract: Full-duplex voice agents need to modulate emotion and delivery during real-time conversations, when de-escalating a complaint, carrying urgency in dispa
Hey everyone, Jovan from UkisAI (Swift Qwen) here! For those who don't know us, UkisAI is a small lab making tiny frontier LLMs, tools and datasets (+doing it open-source!). I'm one of the guys runnin
arXiv:2610.10506v1 Announce Type: cross Abstract: Many of the questions now put to large language models have no correct answer to score against: what a policy is worth, which option a user should cho
Editor’s note: This article is adapted from Thomas Kurian’s keynote address at Gemini at Work 2026. It includes a number of announcements: Gemini, a single, universal agent for work that answers your
arXiv:2610.08680v1 Announce Type: cross Abstract: Endpoint accuracy on abstract-reasoning benchmarks does not reveal whether a language model has acquired a transferable rule or fit distribution-speci
arXiv:2610.07677v1 Announce Type: new Abstract: In this paper, we show that standard evaluations of high-resolution Model Inversion Attacks (MIAs) significantly underestimate training-data privacy lea
arXiv:2508.19073v4 Announce Type: replace-cross Abstract: Deep learning training commonly runs on shared multi-tenant GPU servers, where exclusive allocation provides isolation but can leave resources
Wanted to share a local memory engine I built called Hillock. The main problem I had with standard local RAG was that running document parsing and vector search chewed up so much VRAM that my actual O
arXiv:2610.07249v1 Announce Type: new Abstract: How can an AI agent determine what rules to follow? One rule permits an action. Another imposes a condition, exception, or conflicting obligation. Deter
arXiv:2610.07289v1 Announce Type: cross Abstract: Manual repair of program failures is time-consuming and disruptive for software developers, particularly during the pre-submit phase where test failur
arXiv:2610.07742v1 Announce Type: cross Abstract: Optimized kernels such as FlashAttention and FlashDecoding are crucial for accelerating today's large models. Most of them are handwritten by experts
arXiv:2610.07907v1 Announce Type: new Abstract: Recurrent neural networks typically compress information into a single vector-valued recurrent state, forcing short-term computation and long-term reten
arXiv:2601.16390v2 Announce Type: replace-cross Abstract: Large language models exhibit strong multilingual capabilities, yet significant performance gaps persist between dominant and non-dominant lan
arXiv:2610.08553v1 Announce Type: cross Abstract: Sequential test-time training adapts a memory network through successive updates, each computing an inner-loop gradient based on the network's previou
arXiv:2610.07654v1 Announce Type: cross Abstract: On-policy distillation (OPD) has attracted growing attention as an effective way to transfer capabilities from teacher models to student models. Recen
arXiv:2610.07345v1 Announce Type: cross Abstract: IAM policy analysis typically emphasizes the authorization capabilities encoded in a policy, but security analyst review priority may also depend on t
arXiv:2610.07708v1 Announce Type: new Abstract: Recommender systems learn from observed user-item interactions, but explicit negative feedback is often unavailable. Since deep learning models require
arXiv:2610.08660v1 Announce Type: new Abstract: Background: Biomedical AI can generate plausible explanations without reliably verifying whether each statement is supported by patient-specific evidenc
arXiv:2610.06920v1 Announce Type: cross Abstract: Automatic music annotation is typically tackled under the assumption of a fixed annotation schema. In practice, commercial music catalogs often need t