Knowledge catalogue
全网都在吹的LeCun新论文，90%的解读都是错的。 他们说生成式AI是死路，说过去三年花的几百亿全白费了，说15M参数的小模型就能吊打万亿大模型。 这些全是营销号的夸张， 我觉得这篇论文的真正分量比他们吹的还要重。 Yann LeCun团队这次解决了JEPA困扰了好几年的表征坍缩问题。 以前的世界模型，学着学着就会把狗车人都压成一模一样的向量，什么都学不到。 这次他们只加了一个极其优雅的数学正则
arXiv:2604.17892v1 Announce Type: new Abstract: Recently, latent reasoning has been introduced into large language models (LLMs) to leverage rich information within a continuous space. However, withou
This post celebrates RC Lens fans experiencing an euphoric moment, likely following a significant sporting victory or achievement at their Bollaert stadium. The enthusiastic French-language post with
arXiv:2601.21244v3 Announce Type: replace-cross Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has advanced LLM reasoning, but remains constrained by inefficient exploration under lim
Let's talk parsing charts 📊📈. Last week we released ParseBench, the first document OCR benchmark for AI agents. New in ParseBench: ChartDataPointMatch. Most document look at a chart and OCR the captio
arXiv:2604.17371v1 Announce Type: cross Abstract: This paper investigates communication-efficient neural network transmission by exploiting structured symmetry constraints in convolutional kernels. In
arXiv:2506.00955v2 Announce Type: replace Abstract: Sarcasm fundamentally alters meaning through tone and context, yet detecting it in speech remains a challenge due to data scarcity. In addition, exi
arXiv:2604.16903v1 Announce Type: new Abstract: Collecting embodied interaction data at scale remains costly and difficult due to the limited accessibility of conventional interfaces. We present a gam
arXiv:2512.12643v2 Announce Type: replace Abstract: Legal relations serve as an important analytical framework for dispute resolution in civil cases. However, legal relations in Chinese civil cases re
arXiv:2604.16741v1 Announce Type: new Abstract: Robot navigation in crowded pedestrian environments is a well-known challenge and we explore the practical deployment of group-based representations in
arXiv:2604.16382v1 Announce Type: new Abstract: Longitudinal NLP tasks require reasoning over temporally ordered text to detect persistence and change in human behavior and opinions. However, in-conte
arXiv:2506.00772v2 Announce Type: replace-cross Abstract: Recent studies have shown that supervised fine-tuning of LLMs on a small number of high-quality datasets can yield strong reasoning capabiliti
arXiv:2604.16981v1 Announce Type: cross Abstract: The LEOPs (Light-ERG-Oscillatory Potentials) dataset provides light-adapted (LA) electroretinogram (ERG) and Oscillatory Potentials (OPs) waveforms fo
arXiv:2604.17158v1 Announce Type: cross Abstract: The occurrence of cybersickness in virtual reality (VR) significantly impairs users' perception and sense of immersion. Therefore, timely detection of
arXiv:2601.03043v3 Announce Type: replace Abstract: Large language models (LLMs) demonstrate strong capabilities across a wide range of complex tasks and are increasingly deployed at scale, placing si
arXiv:2604.18199v1 Announce Type: new Abstract: Transformer-based embedding models suffer from quadratic computational and linear memory complexity, limiting their utility for long sequences. We propo
arXiv:2507.10694v2 Announce Type: replace Abstract: Passive deformation due to compliance is a commonly used benefit of soft robots, providing opportunities to achieve robust actuation with few active
arXiv:2604.18274v1 Announce Type: new Abstract: Temporal Action Detection (TAD) in untrimmed videos is currently dominated by Transformer-based architectures. While high-performing, their quadratic co
arXiv:2512.04677v5 Announce Type: replace Abstract: Audio-driven avatar interaction demands real-time, streaming, and infinite-length generation -- capabilities fundamentally at odds with the sequenti
arXiv:2604.17021v1 Announce Type: new Abstract: Video editing aims to modify input videos according to user intent. Recently, end-to-end training methods have garnered widespread attention, constructi
arXiv:2604.17106v1 Announce Type: new Abstract: Motivated by the challenge presented by non-Markovian objectives in reinforcement learning (RL), we present a novel framework to track and represent the
arXiv:2604.04815v2 Announce Type: replace Abstract: The rapid development of Large Language Models (LLMs) has transformed fake news detection and fact-checking tasks from simple classification to comp
arXiv:2602.17036v2 Announce Type: replace-cross Abstract: The continuous expansion of digital learning environments has catalyzed the demand for intelligent systems capable of providing personalized e
arXiv:2507.09025v4 Announce Type: replace Abstract: We propose Lizard, a linearization framework that transforms pretrained Transformer-based Large Language Models (LLMs) into subquadratic architectur
arXiv:2506.03178v2 Announce Type: replace-cross Abstract: Automated radiology report generation holds significant potential to reduce radiologists' workload and enhance diagnostic accuracy. However, g
arXiv:2501.05067v3 Announce Type: replace Abstract: In this paper, we introduce LLaVA-Octopus, a novel video multimodal large language model. LLaVA-Octopus adaptively weights features from different v
arXiv:2604.16555v1 Announce Type: cross Abstract: Neural Architecture Search (NAS) aims to automatically discover high-performing deep neural network (DNN) architectures. However, conventional algorit
arXiv:2603.14937v2 Announce Type: replace-cross Abstract: Text-rich graphs, which integrate complex structural dependencies with abundant textual information, are ubiquitous yet remain challenging for
arXiv:2604.17770v1 Announce Type: new Abstract: Data scarcity remains a fundamental bottleneck in applying deep learning to wireless communication problems, particularly in scenarios where collecting
arXiv:2604.16763v1 Announce Type: new Abstract: Causal inference from electronic health records (EHR) is fundamentally limited by unmeasured confounding: critical clinical states such as frailty, goal
arXiv:2507.02850v3 Announce Type: replace Abstract: We describe a vulnerability in language models (LMs) trained with user feedback, whereby a single user can persistently alter LM knowledge and behav
arXiv:2604.16379v1 Announce Type: cross Abstract: Industrial B2B applications (e.g., construction site risk prediction, material procurement) face extreme data sparsity yet feature rich textual intera
LLMs are still not consistent judges of qualitative work, and small changes to how that work is presented affect outcomes. Better harnessing and methods (multiple judging runs with randomized orders,
arXiv:2604.16935v1 Announce Type: cross Abstract: Scarce longitudinal evidence examines LLMs' persuasiveness and humanness along time-evolving psychological frameworks. We introduce Talk2AI, a longitu
arXiv:2604.17140v1 Announce Type: cross Abstract: We present a generic algorithm for learning and approximate inference with an intuitive epistemic interpretation: iteratively focus on a subset of the
arXiv:2603.24790v2 Announce Type: replace Abstract: While backpropagation and automatic differentiation have driven deep learning's success, the physical limits of chip manufacturing and rising enviro
arXiv:2604.17895v1 Announce Type: new Abstract: Nature suggests that exploiting the elasticities and natural dynamics of robotic systems could increase their locomotion efficiency. Prior work on elast
arXiv:2604.16696v1 Announce Type: new Abstract: 3D object detection in point cloud data remains a challenging task due to the sparsity and lack of global structure inherent in the input. In this work,
arXiv:2601.16504v3 Announce Type: replace Abstract: Commonsense reasoning often involves evaluating multiple plausible interpretations rather than selecting a single atomic answer, yet most benchmarks
arXiv:2604.17346v1 Announce Type: new Abstract: In this book we promote logical computational linguistics as opposed to statistical computational linguistics. In particular, we provide a logical seman
arXiv:2603.26771v2 Announce Type: replace Abstract: Masked diffusion language models (MDLMs) generate text by iteratively unmasking tokens from a fully masked sequence. Their standard confidence-based
arXiv:2510.09354v2 Announce Type: replace Abstract: Large reasoning models exhibit long chain-of-thought reasoning with complex strategies such as backtracking and self-verification. Yet, these capabi
A major European study tracking over 10,000 people aged 65-94 for seven years found that lonely older adults performed worse on initial memory tests, though their memory declined at the same rate as t
arXiv:2604.17428v1 Announce Type: new Abstract: As video generation models achieve unprecedented capabilities, the demand for robust video evaluation metrics becomes increasingly critical. Traditional
arXiv:2604.18258v1 Announce Type: new Abstract: While modern text-to-image (T2I) models excel at generating images from intricate prompts, they struggle to capture the key details when the inputs are
arXiv:2604.16788v1 Announce Type: new Abstract: Robotic manipulation policies often degrade over extended horizons, yet existing benchmarks provide limited insight into why such failures occur. Most p
arXiv:2604.17190v1 Announce Type: new Abstract: Aerial Vision-and-Language Navigation (Aerial VLN) enables unmanned aerial vehicles (UAVs) to follow natural language instructions and navigate complex
arXiv:2511.07129v3 Announce Type: replace Abstract: Low-Rank Adaptation (LoRA) has emerged as a parameter-efficient approach for fine-tuning large language models. However, conventional LoRA adapters
arXiv:2604.18117v1 Announce Type: new Abstract: Post-training quantization (PTQ) is essential for deploying large diffusion transformers on resource-constrained hardware, but aggressive 4-bit quantiza
arXiv:2604.17897v1 Announce Type: new Abstract: The advent of Large Language Models (LLMs) has fundamentally reshaped the way we interact with graphs, giving rise to a new paradigm called GraphLLM. As
arXiv:2604.16836v1 Announce Type: new Abstract: Semantic segmentation in hyperbolic space enables compact modeling of hierarchical structure while providing inherent uncertainty quantification. Prior
Love this work from Aksel and the post-training team at Hugging Face! Turns out the HF ecosystem (papers, datasets, models all accessible through CLI, skills and md files) is perfect for running SOTA
arXiv:2604.17669v1 Announce Type: new Abstract: This paper presents a comprehensive review of the NTIRE 2026 Low Light Image Enhancement Challenge, highlighting the proposed solutions and final result
arXiv:2509.11983v2 Announce Type: replace Abstract: Neural network (NN) training is inherently a large-scale matrix optimization problem, yet the matrix structure of NN parameters has long been overlo
arXiv:2604.16620v1 Announce Type: new Abstract: Analysis of Stochastic Gradient Descent (SGD) and its variants typically relies on the assumption of uniformly bounded variance, a condition that freque
arXiv:2604.18490v1 Announce Type: new Abstract: Existing MT evaluation frameworks, including automatic metrics and human evaluation schemes such as Multidimensional Quality Metrics (MQM), are largely
arXiv:2506.13743v2 Announce Type: replace Abstract: Retrieval-Augmented Generation (RAG) systems typically rely on a single fixed retriever, despite growing evidence that no single retriever performs
LTX 2.3 GGUF 12GB is an updated workflow tool for image generation that now supports multi-image input functionality, with pre-configured workflows for first-frame-last-frame (FFLF) processing using u
arXiv:2604.17976v1 Announce Type: new Abstract: This paper presents ltzGLUE, the first Natural Language Understanding (NLU) benchmark for Luxembourgish (LTZ) based on the popular GLUE benchmark for En
arXiv:2511.09818v2 Announce Type: replace Abstract: Restoring 3D scenes with low-light conditions is challenging, and most existing methods depend on precomputed camera poses and scene-specific optimi