Knowledge catalogue
arXiv:2609.25438v1 Announce Type: new Abstract: Diverse pretraining has been shown to be an effective method for learning reusable, domain-aware representations that provide a starting point for fine-
arXiv:2609.25048v1 Announce Type: new Abstract: How many prompts does on-policy distillation (OPD) need, and how does the answer depend on the student policies that generate its training responses? We
First: That the 7GB (int8) 2.1 comes even close to the 20GB (int8) 2511 is amazing, but maybe because of the smaller size some capabilities got lost. For some tasks 2.1 it is much better than 2511, bu
arXiv:2609.25569v1 Announce Type: new Abstract: Soccer tactics are interactive: an attacking action changes the opponent's defensive problem, and the observed response depends on the multi-agent match
arXiv:2609.26210v1 Announce Type: new Abstract: Vision-language models (VLMs) are increasingly used to interpret charts and generate natural-language explanations for socially consequential data. Howe
arXiv:2609.25009v1 Announce Type: new Abstract: Numerically equivalent word problems should yield the same canonical answer whether a quantity is written as a decimal, fraction, percentage, number wor
arXiv:2609.26083v1 Announce Type: new Abstract: This paper presents a situation aware locomotion framework for two mobile collaborative robots operating in shared industrial environments. The proposed
arXiv:2603.05294v3 Announce Type: replace Abstract: Existing LLM-based web agents struggle on complex, long-horizon tasks due to limited in-context memory, weak planning abilities, and greedy behavior
arXiv:2609.23889v1 Announce Type: cross Abstract: Automated kernel vulnerability reproduction is essential for bug triage, patch validation, and regression testing, but still lacks an effective and ef
arXiv:2609.25657v1 Announce Type: new Abstract: We increasingly use machine learning to label scientific datasets. The models we develop and deploy are improving all the time, but they are not and wil
There’s an enormous amount of open training data on Hugging Face. What does it take to make it work together 🤗? For Marin’s 535B run, we built on 25T tokens from 152 datasets with licenses permitting
Hey everyone, I’m new to running AI models locally and I’d like some advice. My laptop specs are: HP OMEN 15 CPU: Intel i5-8300H RAM: 16 GB GPU: NVIDIA GTX 1050 Ti 4 GB VRAM OS: Windows I’m planning t
Hey everyone, I'm getting into running LLMs locally with Ollama and would like some recommendations on which models I should realistically run with my hardware. My PC specs GPU: RTX 5060 Ti 16GB VRAM
arXiv:2609.25623v1 Announce Type: new Abstract: More privileged information does not always make a better teacher. We study this tension in on-policy self-distillation (OPSD), where a frozen copy of t
I've been comparing and working on Gemma 4, Phi-4-Mini, and Qwen3-4B for local LLM use on an 8 GB Raspberry Pi 5. The results were indeed close, but each model has a different strength: Gemma 4 E2B: a
arXiv:2609.25809v1 Announce Type: new Abstract: Fine-grained mixture-of-experts (MoE) architectures have become a mainstream design for open-weight LLMs, with hundreds of experts and increasingly many
arXiv:2609.25900v1 Announce Type: new Abstract: When we evaluate the performance of our odometry, it is common practice to score the estimated track against a ground truth. Unfortunately, scoring uses
arXiv:2609.24432v1 Announce Type: cross Abstract: Sparse on-policy distillation (OPD) allocates teacher supervision to a small subset of tokens in student-generated trajectories. However, useful teach
arXiv:2609.24328v1 Announce Type: new Abstract: Combining predictions from different models can improve performance at machine learning tasks, but the training of the individual models and the rule us
arXiv:2609.24189v1 Announce Type: new Abstract: Vision-language navigation requires embodied agents to navigate environments using natural language instructions and visual observations. Existing appro
arXiv:2609.24127v1 Announce Type: new Abstract: Atomic activity understanding aims to recognize and localize structured traffic behaviors that jointly encode motion patterns and their grounding in roa
arXiv:2609.24124v1 Announce Type: new Abstract: Active perception and manipulation are crucial for robots to interact with complex scenes. Existing benchmarks struggle to evaluate how robots effective
arXiv:2609.22951v1 Announce Type: cross Abstract: Enterprise agentic systems that route every trajectory step to a frontier model waste 60-80% of their inference budget on subtasks that smaller models
arXiv:2609.24487v1 Announce Type: new Abstract: In this work, we present a method for shape reconstruction and tracking from video via agentic analysis-by-synthesis. Unlike prior methods which first e
Anthropic's Claude Opus 5.5 is now available in Hermes Agent via Nous Portal! OPUS 5.5 BENCHMARKS JUST DROPPED. It beats Fable 5.1 and GPT 6 Astra across the board. Agentic coding, knowledge work, rea
arXiv:2609.23578v1 Announce Type: new Abstract: As AI agents become increasingly capable, agent-driven robotic control is emerging as a compelling paradigm. However, prevailing vision-language-action
arXiv:2609.22592v1 Announce Type: cross Abstract: Training agents with reinforcement learning requires a gym, comprising a task, an executable environment in which the task can be attempted, and a ver
arXiv:2609.24210v1 Announce Type: new Abstract: Building strong chart-to-code systems increasingly relies on reinforcement learning, whose effectiveness depends critically on the quality of the reward
arXiv:2609.23178v1 Announce Type: new Abstract: Language models are appealing tools for research on the past. But to trust the evidence a model provides, researchers need to know whether its responses
arXiv:2609.22813v1 Announce Type: new Abstract: We present commonsense ranked search (CoRS), a novel path planner that turns an abstract instruction into a route that follows commonsense. While existi
arXiv:2609.22415v1 Announce Type: new Abstract: Complex-valued Transformers have inherited softmax attention over the raw complex inner product. Outside natively complex domains this standard form sta
arXiv:2609.22101v1 Announce Type: new Abstract: Large language models can process increasingly long prompts, yet their ability to locate and use decisive evidence may degrade as irrelevant or confusab
arXiv:2609.24749v1 Announce Type: cross Abstract: Latent world models predict the consequences of actions, but accurate prediction does not guarantee that latent distance reflects which candidate will
arXiv:2603.07449v2 Announce Type: replace-cross Abstract: Enterprises commonly deploy heterogeneous database systems, each of which owns a distinct SQL dialect with different syntax rules, built-in fu
arXiv:2507.22920v2 Announce Type: replace Abstract: The rapid advancement of large language models (LLMs) has intensified the need for effective mechanisms to transform continuous multimodal data into
arXiv:2609.22225v1 Announce Type: new Abstract: Large language models (LLMs) exhibit a range of human-like decision-making behaviors, but whether these reflect similar underlying mechanisms or surface
arXiv:2609.24177v1 Announce Type: new Abstract: Legal information in Bangladesh is inaccessible to most citizens. Statutory text is English-only, trained lawyers are concentrated in urban centres, and
arXiv:2609.24746v1 Announce Type: new Abstract: Existing approaches to solving differential equations, such as symbolic regression, physics informed neural networks, and neural operators, typically fo
arXiv:2609.22119v1 Announce Type: new Abstract: Evaluation awareness poses an unprecedented threat to model evaluation, but the mechanisms by which models detect it remain unknown. This study focuses
arXiv:2609.24631v1 Announce Type: new Abstract: Autonomous parking in nonconvex and narrow environments remains challenging. Although optimal-control methods can explicitly enforce vehicle dynamics an
arXiv:2609.24732v1 Announce Type: new Abstract: Diffusion-weighted imaging (DWI) remains highly vulnerable to subject motion, particularly in time-efficient protocols and in motion-prone populations.
Great paper from Google and colleagues. Trains Text-to-SQL agents using multi-agent RL. (bookmark it) This work proposes DualSQL, which splits Text-to-SQL into two agents, one that links the question
arXiv:2511.04473v3 Announce Type: replace-cross Abstract: Retrieval of information from graph-structured knowledge bases represents a promising direction for improving the factuality of LLMs. While va
arXiv:2609.22241v1 Announce Type: new Abstract: We present H2LooP Telecom Model v1, a domain-specialized large language models fine-tuned for the telecommunications industry. We release two domain-ada
arXiv:2506.08296v3 Announce Type: replace Abstract: Recent advances in multimodal vision-language-action (VLA) models have revolutionized traditional robot learning, enabling systems to interpret visi
arXiv:2609.23431v1 Announce Type: new Abstract: Human-object interaction (HOI) detection requires grounding an interacting human-object pair and recognizing the verb that links them, often under sever
In our evaluations, we see that GPT-6 Sol reads about 17% less context than GPT-5.6 Sol to reach the same score, and rarely leaves a repo with failing tests after its last edit. GPT-6 Luna’s biggest g
I am trying out a few fine-tunes of Qwen 3.5 9B @ IQ4_XS @ 131K context and trying to go mostly local (free beats cheap, after all). However, it is much slower than at, say 16K context, even when I am
arXiv:2609.22262v1 Announce Type: cross Abstract: Medical time series (MedTS), including electrocardiograms (ECG), electroencephalograms (EEG), photoplethysmography (PPG), and vital-sign recordings, a
Release: llm 0.36 New OpenAI models: gpt-6-sol for GPT-6 Sol and gpt-6-luna for GPT-6 Luna. #1702 Model plugins can now declare supports_conversation = False for models that only accept single-turn pr
arXiv:2609.23367v1 Announce Type: cross Abstract: FORM is a domain-specific symbolic manipulation language widely used in particle physics for processing the very large algebraic expressions arising f
arXiv:2609.23193v1 Announce Type: cross Abstract: As Large Language Model (LLM) APIs become increasingly integrated into privacy-sensitive workflows, ensuring inference-time privacy without compromisi
arXiv:2609.23312v1 Announce Type: new Abstract: In environments with large movable obstacles, detour-only navigation can be inefficient or even infeasible, while obstacle interaction requires reasonin
arXiv:2609.24526v1 Announce Type: new Abstract: Physical AI requires models to ground visual and linguistic understanding in real-world environments while accounting for environmental constraints and
arXiv:2609.22778v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) are increasingly used as evaluators, yet their reliability in professional assessment tasks that require expert
arXiv:2609.23553v1 Announce Type: new Abstract: The SYNTAX score is a clinically established tool for assessing anatomical lesion complexity in coronary artery disease and guiding subsequent treatment
arXiv:2602.09765v2 Announce Type: replace Abstract: Previous Vision-Language-Action models face critical limitations in navigation: scarce, diverse data from labor-intensive collection and static repr
arXiv:2609.23087v1 Announce Type: new Abstract: Modern Transformer design and compression both reduce to allocating capacity under a budget. The standard scalars for these decisions, #Params and #FLOP
arXiv:2603.05487v2 Announce Type: replace Abstract: Vision-Language-Action models (VLAs) have shown remarkable progress towards embodied intelligence. While their architecture partially resembles that
arXiv:2609.23088v1 Announce Type: new Abstract: Educational foundation models must solve problems, understand curriculum structure, diagnose learner difficulties, and provide appropriate instructional