SignTrace: Describe a Sign, Find the Word
arXiv:2609.30295v1 Announce Type: cross Abstract: Identifying an unfamiliar sign is difficult when a learner remembers its movement but does not know its meaning or formal feature codes. SignTrace add
Knowledge catalogue
arXiv:2609.30295v1 Announce Type: cross Abstract: Identifying an unfamiliar sign is difficult when a learner remembers its movement but does not know its meaning or formal feature codes. SignTrace add
I know most people here probably don't care as much about sonnet 5.5 as they did opus 5.5, because their 100 and 200 plans already give them plenty of weekly usage for opus. But keep in mind: free use
arXiv:2609.30761v1 Announce Type: new Abstract: Most existing human motion generation (HMG) methods use cross-attention modules to inject text semantics, but ignore the importance of bidirectional mod
arXiv:2609.31371v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated promising performance in log anomaly detection, yet how their adaptation strategies, architectures, and d
arXiv:2609.31193v1 Announce Type: new Abstract: Current Audio-Visual LLMs (AVLLMs) struggle with reasoning over videos featuring multi-speaker dialogues. In such videos, resolving 'who says what' is c
arXiv:2609.31293v1 Announce Type: cross Abstract: Speech-based screening is a promising, non-invasive approach for detecting Alzheimer's disease and related cognitive risks. However, models trained on
arXiv:2609.31002v1 Announce Type: new Abstract: Open rerankers trained for general web retrieval transfer imperfectly to e-commerce, where ranking decisions depend not only on topical relevance but al
I run my own LLM while also having a Openai subscription. Also tried DeepSeek (latest flash now). I run Qwen 3.8 flash Next at an amazing speed on my 2x3090 + Ram! But local LLM never did worked for m
I am new to all this so still a lot to learn. If I give Qwen3.8 27B a coding task to fix some bugs in some code, it went through and found and fixed several in like 5 or 10 minutes. I gave qwen flash
EDIT: About that, I tried it out, and it's garbage so far. I did some basic tests through AIHubMix (do not use that platform btw, it's trash), and my agent did some comparison. I guess it figures, it
I know Qwen is a key player in the local LLM space and has consistently introduced truly impactful technologies—like n-gram in Qwen-Next and the recent Qwen 3.8 27B, which is an amazing local model. H
We built an inference engine for MoE models that don't fit in VRAM + RAM. Most of the model stays on the SSD, and experts are read as tokens need them. This started as a proof of concept, and poc work
Sometimes it’s easy to forget that this sub and others like it are probably the extreme minority when it comes to this hobby. Most people, I would think, don’t use or can’t afford one good GPU, let al
I maintain TensorSharp, an open-source inference engine. It can now run Qwen-Image 2.1 locally for text-to-image generation and image editing, with support for its LoRA adapters. I’ve added configs fo
arXiv:2609.28609v1 Announce Type: new Abstract: Role-playing agents based on large language models have been widely applied in areas such as personalized assistance and social simulation. Recent RL me
arXiv:2609.28879v1 Announce Type: cross Abstract: Audio-language models can produce confident answers unsupported by the audio, motivating uncertainty estimates that identify unreliable responses. We
edit: had to change formatting as I’m on mobile noe and the table broke.. I let Codex do a quick eval on Qwen 3.8 Flash Next IQ4_XS (served via vLLM + r9v) and Qwen 3.8 27B FP8 (served via vLLM + Radi
arXiv:2609.29948v1 Announce Type: new Abstract: Prompt injection can degrade benign task performance without eliciting harmful content. Yet many attack objectives depend on task labels or predefined t
arXiv:2609.29006v1 Announce Type: new Abstract: Existing video deraining methods typically exploit neighboring frames through either explicit alignment or implicit spatiotemporal aggregation. Explicit
arXiv:2609.30123v1 Announce Type: new Abstract: Agent skills provide reusable knowledge and instructions, yet agents must repeatedly infer how to apply them and which operation should follow. This cou
arXiv:2512.13130v2 Announce Type: replace Abstract: High-resolution phenotyping at the level of individual leaves offers fine-grained insights into plant development and stress responses. However, the
arXiv:2609.29413v1 Announce Type: new Abstract: Sampling from Boltzmann distributions of molecular systems is an inference problem that has seen significant recent developments fuelled by advances in
arXiv:2609.29718v1 Announce Type: cross Abstract: Coding agents are beginning to act in the visual world. They now build webpages, GUIs, games, 3D scenes, diagrams, and documents. Success in such visu
arXiv:2609.29604v1 Announce Type: new Abstract: In medical visual question answering (VQA), hallucinations of vision-language models (VLMs) may lead to confident but incorrect responses, raising the r
I’ve been experimenting with whether Qwen3.8-Flash-Next’s pretrained PLE n-gram memory can improve a much smaller Qwen3.5-0.8B model. I trained the 0.8B setup with limited resources, mostly using free
arXiv:2609.30219v1 Announce Type: cross Abstract: An acceptance protocol is developed for sensor-coordinate and polarity binding in mechatronic commissioning. Candidate generation is separated from re
arXiv:2609.29715v1 Announce Type: cross Abstract: Surrogate models are often chosen during development and then left in place as new measurements arrive. That practice becomes risky when noise, input
arXiv:2609.29543v1 Announce Type: new Abstract: Agent skills bundle procedural guidance with execution conditions governing authority, user consent, and live environment state. When model capabilities
FYI the engine's default kernel rules were measured on smaller chips (16/20-core M5s and a 32-core M4 Max), so a 40-core M5 Max runs guesses. Splash's repo includes a developer tool, “make tune-kernel
arXiv:2609.29549v1 Announce Type: new Abstract: Post-training pipelines must select one language-model policy from many checkpoints, prompts, and decoding rules. Mean evaluator scores can conceal rare
arXiv:2609.30227v1 Announce Type: cross Abstract: Online misinformation increasingly appears in spoken formats such as news clips, podcasts, interviews, political speeches, and social media videos, cr
I have a tool that uses Gemini Flash with the lowest thinking budget to do summarization work. It's very fast, 0.9-1.2s in most cases. But I have a user experience problem where people make the wrong
arXiv:2609.28154v1 Announce Type: new Abstract: Global building and settlement datasets increasingly support population mapping, exposure assessment, urban monitoring, and other analyses of the built
arXiv:2609.28416v1 Announce Type: cross Abstract: Recent advances in large language models (LLMs) have enabled agents to tackle long-horizon tasks across diverse environments. To further improve agent
arXiv:2512.04125v2 Announce Type: replace Abstract: Large language models (LLMs) have demonstrated several emergent behaviors with scale, including reasoning and fluency in long-form text generation.
arXiv:2609.27359v1 Announce Type: new Abstract: Vietnam's Personal Data Protection Law (Law No. 91/2025/QH15) and Decree No. 356/2025/ND-CP, effective January 1, 2026, require organizations to establi
arXiv:2609.27275v1 Announce Type: new Abstract: Most autonomous-driving datasets record only the action executed by a behavior policy and the single future that followed, providing limited supervision
arXiv:2609.27421v1 Announce Type: new Abstract: Multi-constraint instruction following requires a model to respond to a query under many simultaneously active constraints. Even strong instruction-tune
Decision models should do more than pick a label. To truly act, they need to extract evidence, understand relationships, produce structured records, and keep related decisions consistent. And they sho
arXiv:2609.27470v1 Announce Type: cross Abstract: Recent video-language models increasingly adopt hybrid architectures that interleave linear and full attention layers for efficient long-context proce
arXiv:2609.27784v1 Announce Type: cross Abstract: Semi-structured documents are ubiquitous in scientific reports, financial statements, and technical manuals. Question answering over such documents re
arXiv:2609.27308v1 Announce Type: new Abstract: We study coding agents for long-horizon, dexterous robotics and ask whether their solutions can provide scalable supervision for learning general robot
arXiv:2609.27654v1 Announce Type: cross Abstract: Verified income is often unavailable in digital loan applications, forcing lenders to rely on reported income and potentially leading to over-lending,
arXiv:2609.28395v1 Announce Type: new Abstract: Fine-tuning large language models on parallel data improves translation quality but can cause catastrophic forgetting. Mitigation methods are generally
arXiv:2609.27156v1 Announce Type: new Abstract: Large reasoning models can produce correct yet unnecessarily long reasoning traces. Existing methods improve reasoning efficiency with trajectory-level
Impressive paper from Salesforce. It discusses the importance of good verifiers for RL environments. Only 35.8% of the environments in the cleanest public RL collection for terminal agents passed Sale
arXiv:2609.27473v1 Announce Type: new Abstract: PPG-to-vital-sign reconstruction turns a wrist-worn photoplethysmogram into clinical waveforms such as the ECG. Long-horizon multivariate time-series fo
arXiv:2608.24707v2 Announce Type: replace Abstract: While text-based hallucination detection is well studied, reference-free detection of factual alterations in speech remains underexplored, especiall
arXiv:2609.28344v1 Announce Type: cross Abstract: Audio-language models (ALMs) integrate acoustic perception with the knowledge encoded in language models, enabling contextual understanding of auditor
I'm in a position where I didn't mean to go down this AI rabbit hole... I was looking into mental cognitions and how the mind is made up and learned that my old RTX 5k could run small models. So that
Just a quick PSA. llama.cpp does have prompt caching. if you are running large context lengths and have long multiturn projects, increasing -cram can provide you with massive speedups. There is a poin
arXiv:2609.27855v1 Announce Type: new Abstract: AI systems increasingly claim to optimize prompts, policies, architectures, plans, tool-use trajectories, reasoning traces, and test-time computation. T
arXiv:2609.27173v1 Announce Type: new Abstract: Solving complex tasks in domains such as science, medicine, law, and finance often requires assembling interdependent information scattered across vast,
arXiv:2609.27938v1 Announce Type: new Abstract: Enabling operators to perceive and interact with remote environments naturally is a fundamental challenge in robotic teleoperation. This is especially c
arXiv:2609.27442v1 Announce Type: new Abstract: 3D reconstruction from satellite imagery is essential for large-scale topographic analysis, yet the lack of high-fidelity training datasets with accurat
arXiv:2609.27175v1 Announce Type: cross Abstract: Multimedia verification requires not only accurate decisions but also traceable evidence, reliable human correction, and safe reuse of prior experienc
Been building some multi-agent pipelines recently and the degradation in direct execution is insane. Whenever I pass a strict operational variable or hypothesis, the LLM triggers this 'Devil's Advocat
arXiv:2609.25804v2 Announce Type: replace Abstract: LLM agents increasingly work on long-horizon tasks, and the decisions they make along the way, such as which hypothesis to test or which implementat
arXiv:2609.26830v1 Announce Type: cross Abstract: Simulation success is not equivalent to structural correctness for LLM-generated circuits. We define and measure four evaluation levels -- schema vali
arXiv:2609.27822v1 Announce Type: cross Abstract: A common multi-agent design asks agents to report confidence and lets the highest-scoring agent speak next, implicitly using one scalar both to route