Knowledge catalogue
arXiv:2610.11773v1 Announce Type: new Abstract: Guard models are increasingly used to safeguard LLM-based agents, primarily by identifying actions that agents are forbidden to perform. However, identi
arXiv:2610.10625v1 Announce Type: cross Abstract: Looped Language Models (LoopLMs) provide a parameter efficient approach to scaling model capabilities through repeated use of shared parameters across
arXiv:2610.11552v1 Announce Type: new Abstract: Large language model agents can invoke tools fluently, but enterprise workflows demand more than selecting the right tools: actions must strictly comply
arXiv:2610.10627v1 Announce Type: new Abstract: Deep reinforcement learning has achieved substantial performance gains over classical control approaches. Yet, a central challenge to learning in real-w
arXiv:2610.10624v1 Announce Type: cross Abstract: Spatial audio-visual understanding requires models to recognize not only what is present, but also where events occur and how they relate across modal
arXiv:2610.11570v1 Announce Type: new Abstract: In this paper, we argue that looped Transformers need their own residual connections to prevent performance degradation as the number of iterations grow
arXiv:2610.12207v1 Announce Type: new Abstract: Pre-trained transformer models are increasingly being used to study scientific and technological progress. Encoders tuned to paper or patent text outper
arXiv:2610.11583v1 Announce Type: cross Abstract: End-to-end autonomous driving demands trajectory planners that are both highly accurate and cheap enough for edge deployment. State-of-the-art artific
arXiv:2610.12409v1 Announce Type: new Abstract: Safety benchmarks typically report one overall score for a suite of datasets, each of which may target one or more safety-related attributes, so models
arXiv:2610.10703v1 Announce Type: new Abstract: Eyeglass reflection removal is important across smartphone imaging, video conferencing, and other face-centric visual applications. The task is challeng
arXiv:2610.10857v1 Announce Type: new Abstract: Behavior cloning (BC) in non-Markovian environments is a challenging problem because policies have to reason over contextual information over long horiz
arXiv:2610.11567v1 Announce Type: cross Abstract: Battery state of health (SoH) forecasting is important for battery management, but remains challenging due to nonlinear degradation and heterogeneity
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
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:2610.11575v1 Announce Type: new Abstract: Sparse Mixture-of-Experts models scale parameter capacity efficiently while maintaining a fixed compute budget per token. However, traditional training
arXiv:2610.12402v1 Announce Type: cross Abstract: Existing spatial reasoning benchmarks mainly test spatial perception: reading off relations already visible in the input. Yet real-world spatial intel
arXiv:2610.12327v1 Announce Type: cross Abstract: The memory-bound nature of the decoding stage of large language model (LLM) inference incurs significant latency. Layer-wise training-free network pru
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:2610.10965v1 Announce Type: new Abstract: The Muon optimizer orthogonalizes each update matrix, setting all of its singular values to one, and has proven highly effective for training large lang
arXiv:2610.10571v1 Announce Type: new Abstract: Electroencephalography (EEG) provides a non-invasive measure of ongoing neural activity, but building general-purpose EEG models remains challenging due
arXiv:2610.11456v1 Announce Type: new Abstract: Two-photon calcium imaging is a standard tool for recording large neural populations in vivo, yet inferring spikes accurately across the growing diversi
arXiv:2610.11076v1 Announce Type: cross Abstract: Domain-specific continual adaptation of LLMs risks catastrophic forgetting, creating a fundamental tension between acquiring new capabilities and pres
arXiv:2610.11943v1 Announce Type: new Abstract: Autonomous rovers navigating large unstructured environments need efficient global planning that accounts for terrain traversability. However, searching
arXiv:2610.12201v1 Announce Type: cross Abstract: Full-duplex speech models can listen and speak simultaneously, enabling natural interaction, but become increasingly difficult to control as the conve
All anyone ever talks in this sub is about x new model having x benchmark scores and how well it can code. Or some new tech to increase t/s for qwen. So I wanted to go back to the roots of this sub an
arXiv:2610.10878v1 Announce Type: cross Abstract: On-policy distillation (OPD) efficiently transfers capabilities from a stronger teacher to a student language model through dense token-level supervis
arXiv:2610.11496v1 Announce Type: new Abstract: Pose-guided diffusion models can now synthesize entire human figures in motion, spawning a new class of deepfakes: Motion Aware Deepfake (MAD) that have
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:2610.10942v1 Announce Type: new Abstract: Reinforcement learning environments are now a primary lever for improving large language model (LLM) capabilities in post-training, yet most agentic ben
arXiv:2610.10560v1 Announce Type: cross Abstract: AI foundation models pretrained on weather and climate data are increasingly fine-tuned to Earth science tasks well beyond weather forecasting. Their
arXiv:2610.10768v1 Announce Type: cross Abstract: The global challenge of climate change has driven significant steps to reduce CO2 emissions, guided by international agreements like the Paris Agreeme
A bit of backstory : i got lucky in August and was able to grab a CMP 170 HX 8GB for around 500 on alibaba. I tried later to get a second one for around 1k but my order got cancelled and prices went t
arXiv:2610.11579v1 Announce Type: new Abstract: To adapt billion-parameter Vision Transformers for long-video understanding, recent methods freeze the backbone and train lightweight convolutional modu
arXiv:2610.11559v1 Announce Type: cross Abstract: Coding assistants such as Claude Code and Codex have become a major application of LLM agents, yet existing benchmarks remain far from real-world use,
arXiv:2610.10807v1 Announce Type: new Abstract: Estimating the symbolic or analytical form of probability density functions (PDFs) from observed samples is a fundamental challenge in statistical and c
arXiv:2610.12256v1 Announce Type: cross Abstract: Omnimodal embeddings naturally involve both shared representations and modality-specific features across heterogeneous inputs. However, existing omnim
arXiv:2610.11952v1 Announce Type: new Abstract: Instruction following enables robots to perform diverse tasks specified in natural language, making it a fundamental capability for human-robot interact
Youtu-Parsing-Omni is a compact (5B) omni-modal parsing model. Given a single input — a document page, a natural image, a chart / flowchart, a geometry figure, an audio clip or an audio-visual video —
arXiv:2610.12005v1 Announce Type: new Abstract: Which forms of test-time compute improve the predictions of strong pretrained tabular foundation models (TFMs)? We systematically study this along three
I tested Mellum2.1-12B-A2.5B (Q8) locally using Pi and llama-server. All five tests were one-shot. Results were pretty mixed: - Pelican SVG, Browser OS, Minecraft: Poor results. - Bouncing Hexagon: Ph
arXiv:2610.10619v1 Announce Type: cross Abstract: Large language model (LLM) agents are rapidly reshaping software engineering, accompanied by an explosion of new code benchmarks. Yet nearly all exist
The challenge for Google post-Gemini 4 will be what they do with a good model. Anthropic & OpenAI show we are moving towards a single interface for many tasks with orchestrator agents. The Gemini 3 er
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
arXiv:2610.11105v1 Announce Type: new Abstract: Generating high-quality and controllable camera motion is essential for AI-assisted cinematography, video synthesis, and 3D scene understanding. We intr
arXiv:2610.12242v1 Announce Type: new Abstract: Large language model (LLM) routing distributes inference work across different models, advancing the cost-quality Pareto frontier of LLM serving. While
arXiv:2610.11736v1 Announce Type: new Abstract: Modern video generators can realize increasingly complex visual narratives, positioning the prompt enhancer (PE) as a critical bridge from concise user
arXiv:2610.11678v1 Announce Type: new Abstract: Verifier scores now serve as both benchmark metrics and training rewards for large language model (LLM) agents, and a change in score is routinely read
arXiv:2610.12232v1 Announce Type: new Abstract: Test-time adaptation (TTA) methods for time-series forecasting update a deployed model, or a small adapter around it, from incoming ground truth. But th
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
Release: ttok 1.0 I released ttok 0.4, ran uv tool upgrade ttok, piped a file into the new version... and realized that it was defaulting to the GPT-4 tokenizer when it should very clearly default to
arXiv:2610.11392v1 Announce Type: new Abstract: System One models output calibrated probabilities over typed answers such as categorical choices, ordinal levels, or binary outcomes, via a non-generati
arXiv:2610.11764v1 Announce Type: new Abstract: Reliable crop-row perception is essential for autonomous agricultural robots, but low-cost deployment is constrained by computation, memory, power, and
arXiv:2610.11737v1 Announce Type: cross Abstract: Bayesian physics-informed neural networks (B-PINNs) are a popular framework for parameter and state inference from sparse or noisy observations. They
arXiv:2610.12114v1 Announce Type: new Abstract: Knowledge distillation (KD) transfers knowledge from stronger Teacher models to weaker Student models, but most methods require training the Student par
arXiv:2610.12124v1 Announce Type: new Abstract: The evolution of Large Language Model agents from single-task execution to long-term autonomous operation highlights the critical challenge of transform
arXiv:2609.32119v2 Announce Type: replace Abstract: Language models (LMs) are often used as a tool to model human language processing. Recent studies suggest that severely restricting LMs' context win
arXiv:2602.24230v2 Announce Type: replace-cross Abstract: In probabilistic classification, calibration error (CE) measures the average divergence of predicted probabilities f(X) from P(Y|f(X)), the tr
arXiv:2610.12229v1 Announce Type: cross Abstract: In text-guided image editing, describing the desired change is often straightforward, but identifying the intended object or region can be cumbersome,
arXiv:2610.12435v1 Announce Type: cross Abstract: Teaching a humanoid to follow instructions with its whole body runs into two obstacles. Its action space is large and tightly coupled: legs, arms, and
arXiv:2610.10639v1 Announce Type: cross Abstract: Visible Chain-of-Thought (CoT) is often treated as a broadly useful reasoning instruction, yet analytics code generation combines natural-language amb