Knowledge catalogue
arXiv:2607.20410v1 Announce Type: new Abstract: Value alignment of Large Language Models (LLMs) has been shown to be culturally biased toward Western norms. This results in the mishandling of local va
The math behind reinforcement learning (RL) post-training for large language models (LLMs) is notoriously unforgiving. As frontier AI labs push the boundaries of reasoning and coding models using RL p
arXiv:2607.17820v2 Announce Type: replace Abstract: Training-free few-shot adaptation methods have gained significant attention recently in the context of Vision-language Models (VLMs). Yet, current b
arXiv:2607.20058v1 Announce Type: new Abstract: Large language models can answer scientific questions, yet a correct output does not reveal whether the model represents or uses the governing physics.
arXiv:2607.18709v2 Announce Type: replace Abstract: Existing robot datasets remain expensive to curate, embodiment-specific, and insufficiently annotated with the fine-grained structure required for g
arXiv:2607.19620v1 Announce Type: cross Abstract: In this paper, we present SCPP (Soft Clustering Python Package), an open-source Python framework for soft clustering. SCPP establishes a canonical, sc
arXiv:2607.19404v1 Announce Type: cross Abstract: Multivariate time series encode structural patterns that unfold across multiple temporal scales, yet most forecasting backbones treat learned represen
arXiv:2607.18907v1 Announce Type: new Abstract: Instance-level artwork recognition requires matching a handheld visitor photograph to a specific work in a large museum collection. This is challenging
arXiv:2607.17191v2 Announce Type: replace Abstract: Human-like private chat requires more than fluent response generation: a system must preserve persona, relationship, memory, bounded knowledge, medi
arXiv:2607.18798v1 Announce Type: new Abstract: Reconstructing high-fidelity facial geometry with an assigned topology is essential for digital avatar creation and animation, yet existing automated me
New research from Meta. (bookmark it) Most factuality work checks whether the claims in an answer are correct. GAMUT goes after the harder question of whether the answer covers everything it should. I
BREAKING: Kimi K3 by @Kimi_Moonshot is 1st overall on 3D Design with an Elo of 1450. This is a 6 position and 108 Elo jump from @Kimi_Moonshot's previous model, Kimi K2.6. This performance puts Kimi K
feeling nostalgic, my favorite blogs on RL & reward hacking https://www.alexirpan.com/2018/02/14/rl-hard.html https://lilianweng.github.io/posts/2024-11-28-reward-hacking/ TLDR: An openai model, durin
Last Week in AI #250 is the podcast’s 250th episode, summarizing recent frontier‑AI policy developments. It reports that the U.S. government has granted Anthropic permission to release Mythos‑5 to a l
My 2hr workshop on Open vs Closed models, reward hacking, benchmaxxing & RL is out! 1. Closed vs open models 2. Throughput maxxing but accuracy minimizing 3. Benchmaxxing & cheating 4. Distillation &
I am currently running Qwen3.6-30B in Ollama, through Cline to use as an agent in VSCode. Qwen's skill in coding is not in question, but the performance in VSCode is slow and inaccurate and times out
arXiv:2607.13940v1 Announce Type: new Abstract: Personal health management unfolds over repeated encounters, yet most health AI systems treat each request in isolation. We developed HealthClaw, an ope
arXiv:2607.11707v2 Announce Type: replace-cross Abstract: Following the rapid progress of generative Artificial Intelligence, there is a growing threat posed by conversational scams. These scams often
arXiv:2510.04876v3 Announce Type: replace Abstract: Benthic habitat mapping is fundamental for understanding marine ecosystems, guiding conservation efforts, and supporting sustainable resource manage
arXiv:2607.13643v1 Announce Type: cross Abstract: Sampling multiple solutions and returning the majority answer is among the most reliable ways to improve the reasoning accuracy of large language mode
arXiv:2607.13792v1 Announce Type: new Abstract: Most daily activities are inherently procedural. However, existing evaluations for egocentric video understanding seldom address procedural understandin
arXiv:2607.13303v1 Announce Type: cross Abstract: Formal contracts are essential for software testing and verification, yet writing them remains labor-intensive and error-prone. LLMs offer a promising
arXiv:2607.13587v1 Announce Type: cross Abstract: Automatic symbolic music analysis has made substantial progress, yet existing systems are typically designed for a single mode of use, such as full-sc
arXiv:2607.13080v1 Announce Type: cross Abstract: Autonomous coding agents force engineering organizations to choose between API-based frontier models -- strong reasoning at high token cost -- and on-
arXiv:2601.20496v2 Announce Type: replace-cross Abstract: Generating dense physical fields from sparse measurements is a fundamental question in sampling, signal processing, and many other application
arXiv:2606.16667v2 Announce Type: replace Abstract: Large vision-language models (LVLMs) hallucinate: they assert visual details that the image does not support. A principled remedy is selective predi
arXiv:2607.13639v1 Announce Type: cross Abstract: We introduce OvisOCR2, a 0.8B document parsing model. OvisOCR2 is designed as an end-to-end parser: given a document page image, it generates a Markdo
arXiv:2607.13941v1 Announce Type: new Abstract: Video aesthetic assessment (VAA) aims to predict how aesthetically pleasing a video is, yet remains far less explored than other visual assessment tasks
arXiv:2607.13091v1 Announce Type: cross Abstract: LLM-based coding agents repeat the same classes of mistakes across sessions because they lack a mechanism to retain corrections from human review feed
arXiv:2607.13681v1 Announce Type: new Abstract: Humans can efficiently parse continuous sensory streams, from hours to years, scaffolding an internal world model that grounds spatial reasoning and pre
arXiv:2607.13988v1 Announce Type: new Abstract: Multi-turn agents solve complex tasks through extended sequences of tool interactions before producing a final answer, making credit assignment a fundam
arXiv:2607.13099v1 Announce Type: cross Abstract: Large language models (LLMs) have achieved remarkable success but raise growing concerns about content provenance and misuse, motivating the need for
arXiv:2607.12176v1 Announce Type: new Abstract: Ambivalence and hesitancy (A/H) undermine digital behaviour-change interventions, and recognizing them automatically from video is the goal of the ABAW
arXiv:2607.10383v2 Announce Type: replace-cross Abstract: Visual Language Navigation foundation models aim to unify deep reasoning for grounded spatial decisions with broad versatility for diverse emb
arXiv:2607.12829v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) offer a theoretical advantage in parallel generation over standard autoregressive models. However, parallel ge
arXiv:2607.12469v1 Announce Type: cross Abstract: Many agent-safety evaluation results are not yet load-bearing evidence: identical nominal outcomes (task success, attack success, monitor scores) may
arXiv:2607.12051v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly being explored for clinical decision support, but their reliability in complex oncology treatment planning
LLMs handle speech well once you run speech-to-text. They don't hear the rest: a bird outside, a glass breaking two rooms away, a smoke alarm two floors down. I've been working on an experimental open
arXiv:2607.12058v1 Announce Type: cross Abstract: Given a vulnerability-fixing commit, trigger localization asks which specific statement turns the vulnerable program state into a concrete unsafe oper
Below Upstream Status sections are from https://github.com/PrismML-Eng/Bonsai-demo Upstream Status for Binary Q1_0 is supported out of the box in upstream llama.cpp across many backends: CPU (generic,
arXiv:2607.12112v1 Announce Type: cross Abstract: Federated fine-tuning of Multimodal Large Language Models (MLLMs) across distributed networks enables privacy-sensitive adaptation to evolving data st
arXiv:2602.05513v3 Announce Type: replace-cross Abstract: Bimanual dexterous manipulation relies on integrating multimodal inputs to perform complex real-world tasks. To address the challenges of effe
arXiv:2607.12265v1 Announce Type: new Abstract: Radar sensing is increasingly used in mobile systems because it operates reliably under poor lighting, adverse weather, and privacy-sensitive settings w
arXiv:2607.13034v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly automate multi-step engineering and informatics workflows, yet they rarely ask how much effort a task act
arXiv:2607.12336v1 Announce Type: cross Abstract: Artificial Intelligence (AI) technologies, while serving as a foundational enabler for modern social media and digital health services, exert a bivale
arXiv:2607.12204v1 Announce Type: new Abstract: Attention can be viewed as an online learner over context, yet existing test-time memories cannot certify that dropping a token leaves outputs unchanged
arXiv:2607.12982v1 Announce Type: new Abstract: Math reasoning has achieved significant progress with the rapid advancement of Multimodal Large Language Models (MLLMs), however analytic geometry remai
arXiv:2602.06486v4 Announce Type: replace Abstract: Evaluating agentic AI on open-ended professional tasks faces a fundamental dilemma between rigor and flexibility. Static rubrics provide rigorous, r
arXiv:2607.12704v1 Announce Type: new Abstract: Multi-label classification assigns several co-occurring labels to each aerial scene, yet deployed models often encounter data distributions different fr
What it is: ATH-MaaS/OvisOCR2 - a 0.8B document-parsing VLM post-trained from Qwen3.5-0.8B (SFT + RL + OPD), Apache 2.0, runs on vLLM 0.22.1. One prompt per page image -> complete markdown (HTML table
arXiv:2607.12495v1 Announce Type: new Abstract: Creating photorealistic 3D assets requires bridging the appearance gap between real-world observations and synthetic models. A promising approach is to
arXiv:2607.11193v2 Announce Type: replace-cross Abstract: To ensure the overall quality of AI-enabled software, not only traditional software components but also AI components need to be tested and re
arXiv:2607.12488v1 Announce Type: new Abstract: Molecular optimization in drug discovery, materials design, and catalysis requires searching vast chemical spaces under tight evaluation budgets, since
arXiv:2607.11950v1 Announce Type: cross Abstract: Brain field potentials are scale-free: their power spectra follow a 1/f^{eta} law whose aperiodic exponent eta tracks cortical state, and sleep depth
arXiv:2607.11920v1 Announce Type: cross Abstract: Evaluating decisions made under uncertainty is hard when labeled outcomes are scarce, costly, or confounded with luck. We treat subjective expected ut
Introduction RxBrain (Hy-Embodied-RxBrain-1.0) is a unified multimodal foundation model for embodied cognition — a single model that couples language reasoning with visual imagination to deliver three
arXiv:2607.11933v1 Announce Type: new Abstract: Cross-encoders achieve high reranking accuracy in Retrieval-Augmented Generation (RAG) pipelines but impose quadratic inference costs that limit real-ti
arXiv:2607.12800v1 Announce Type: new Abstract: Learning broad world knowledge directly from raw visual data is a fundamental capability of intelligence. We introduce UniVR, the first investigation in
arXiv:2607.12756v1 Announce Type: new Abstract: Vision-language models (VLMs) process large numbers of visual tokens, resulting in substantial inference latency and memory overhead. This has motivated
arXiv:2602.16918v2 Announce Type: replace-cross Abstract: We present Xray-Visual, a unified vision model architecture for large-scale image and video understanding trained on industry-scale social med