Knowledge catalogue
arXiv:2608.26091v1 Announce Type: cross Abstract: Civil infrastructure compliance checking has long relied on engineers manually reading legacy 2D plans; however, OCR-based automation strips away the
arXiv:2608.25655v1 Announce Type: new Abstract: Conversations with chat assistants increasingly span many topics in a single long-running thread, challenging memory systems. Existing long-context and
arXiv:2608.25973v1 Announce Type: cross Abstract: Understanding instruction-following capabilities in scientific domains is essential for effectively leveraging Multimodal Large Language Models (MLLMs
arXiv:2608.25826v1 Announce Type: new Abstract: A recent line of synthetic-data work reconstructs the thinking behind existing text rather than rewriting the text itself, but it operates on short web
arXiv:2608.26103v1 Announce Type: cross Abstract: Zero-shot cross-task generalization, where a policy must execute manipulation tasks never seen during training, remains a central challenge in robot l
arXiv:2504.18346v4 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have been transformative across many domains. However, hallucination, i.e., confidently outputting incorrect info
arXiv:2608.24228v1 Announce Type: new Abstract: Many LLM applications are most useful when they provide several candidate outputs for comparison, validation, or combination. Predominant evaluation set
arXiv:2608.23962v1 Announce Type: new Abstract: When an LLM serving deployment runs out of KVcache room, there are two well-established ways out. Tensor parallelism shards the weights and the KV cache
arXiv:2601.16520v2 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) have achieved remarkable progress in visual recognition and semantic understanding, yet precise compo
arXiv:2608.23635v1 Announce Type: cross Abstract: Large language models (LLMs) rely on tool calling as a fundamental agent capability, enabling them to invoke external systems and complete tasks beyon
Benchmarked qwen3.8 xhigh, medium and muse glimmer. Xhigh effort mode with qwen3.8 took almost 30hrs. (And still failed on 16 cases because of the 32K output token limit) Medium effort mode and muse g
arXiv:2608.23313v1 Announce Type: new Abstract: Vision-language model safety benchmarks typically evaluate only final responses: whether a model refuses, warns, or complies. This outcome-level view ca
At $10k, you could get - 6.2B tokens with Qwen 3.8 Max (Qwen Pro plan) - 5.7B tokens with DeepSeek V4 Pro OpenRouter - 100B tokens with DeepSeek V4 Flash OpenRouter As a firm believer of local inferen
arXiv:2608.22323v1 Announce Type: new Abstract: The application of Large Language Models (LLMs) to diagnostic decision-making has garnered growing interest. However, existing benchmarks largely focus
arXiv:2608.23474v1 Announce Type: cross Abstract: Vision-language models (VLMs) achieve strong performance on video and image-sequence benchmarks, yet it remains unclear whether they capture temporal
AgentX 1.0 is the first fully open‑source multi‑turn agentic coding inference benchmark (Apache 2.0), released August 2026 with a 1 million‑token context and over 1 million tokens of dataset built at
arXiv:2608.20344v1 Announce Type: new Abstract: LLM-based 'digital twins' aim to simulate how an individual would behavein new environments or respond to novel questions, given some representation of
arXiv:2608.21089v1 Announce Type: new Abstract: Integrating Large Language Models (LLMs) into the Indian judiciary promises access to justice but introduces severe risks. We identify the 'inertia of c
arXiv:2608.20960v1 Announce Type: new Abstract: Language models (LMs) are trained on static scientific corpora, whereas scientific knowledge continuously evolves through correction and revision. Scien
arXiv:2608.20769v1 Announce Type: cross Abstract: Streaming emotion understanding uses historical state while continuously interpreting current audio, often feeding the model's previous prediction bac
arXiv:2608.21021v1 Announce Type: cross Abstract: Real-world fault analysis in 5G and emerging 6G networks demands domain expertise to analyze free-text diagnostics, including root-cause explanations
NVIDIA Groq 3 LPX, when paired with the Vera Rubin NVL72 platform, delivers a record‑setting 3,431 output tokens/second on the Artificial Analysis 100K context benchmark using Gemma 4 31B, demonstrati
arXiv:2608.21160v1 Announce Type: new Abstract: Machines that understand humans should perceive the present and anticipate the future. Existing human-centric vision model are pretrained on human image
arXiv:2608.20840v1 Announce Type: cross Abstract: Recent advances in multimodal retrieval have improved the ability to retrieve information from visually rich documents such as PDFs and reports. Howev
arXiv:2608.20755v1 Announce Type: new Abstract: Modern de novo design workflows generate many candidate protein binders, but wet-lab validation capacity remains limited, making shortlisting a major bo
I ran the Aider benchmark on Qwen 3.8 27B FP8 with FP8 KV cache 256K context vLLM. The score: 72.9 This matches Gemini 2.5 Pro from 2025-04-12 which also scored 72.9. Beats Claude Opus 4 from 2025-05-
Thanks to Unsloth Dynamic 3.0 quants coming in slightly leaner and better preserved, I settled on Qwen 3.8 27B (`UD_Q4_K_XL`) at 100K context as my daily driver for Hermes Agent and OpenCode. On an RT
I'm running Qwen 3.8 27B locally on a single GPU. I can push the context to 131k, but I'd rather run it faster at 64k if the agent can manage context properly. What I have in mind is pretty simple: on
arXiv:2608.19932v1 Announce Type: new Abstract: Public cardiac cohorts annotate different subsets of the heart, so shapes from separate sources cannot be pooled without shared correspondence. Among re
arXiv:2608.20107v1 Announce Type: new Abstract: Recent advances in generative video models have significantly improved visual realism in video object removal, yet evaluation protocols still focus on m
arXiv:2608.19652v1 Announce Type: new Abstract: As LLM-based agents are deployed for longer and higher-stakes tasks, their memory systems continue to have crucial gaps. While existing memory benchmark
Here's a brand new Blackwell-native, prefill-optimized 4-bit quant that runs 50% faster on compatible hardware than a Q4 quant of the same memory footprint. And it runs 4-7% faster than other NVFP4 qu
arXiv:2608.19355v1 Announce Type: cross Abstract: Educational visual question answering, or VQA, requires models to solve curriculum-oriented multiple-choice questions using both language and visual e
Harvey is a great example of how American companies are building world-class specialized models: they took an open-source base (Kimi K3), post-trained it on legal data, and delivered state-of-the-art
arXiv:2608.19746v1 Announce Type: new Abstract: Personalized text generation aims to make LLMs write in a specific individual's style, yet existing benchmarks measure task accuracy or preference align
arXiv:2608.19564v1 Announce Type: new Abstract: Persistent memory can personalize an LLM agent, but an incorrect durable update can silently distort future behavior. We study the memory-clarification
arXiv:2608.20237v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) combine linguistic reasoning with visual perception, yet their ability to perform visual spatial planning under
arXiv:2608.19987v1 Announce Type: new Abstract: Skeleton-based Video Anomaly Detection (VAD) offers a robust, privacy-preserving solution for identifying abnormal behaviors. To model the distribution
arXiv:2608.19825v1 Announce Type: cross Abstract: Medical image captioning is a technique that accelerates early-stage diagnostic workflows and enhances the interpretability of medical diagnostic AI s
arXiv:2608.18903v1 Announce Type: new Abstract: This work evaluates surrogate-assisted optimization of a seven-parameter current-excited coil--core benchmark subject to geometric, manufacturing, and s
Hey LocalLlama, Henry from Cactus here! When we trained Needle 2, I had a strict rule to not expose the model to any data sample that remotely felt like these benchmarks. It seemed over-the-top, but b
arXiv:2608.18099v1 Announce Type: new Abstract: Investment management is a high-stakes domain in which agentic AI systems must do more than generate plausible text. They must retrieve point-in-time da
arXiv:2608.18993v1 Announce Type: new Abstract: Electronic travel aids are pivotal for the independent mobility of the visually impaired. While Vision-Language Models (VLMs) offer rich environmental u
arXiv:2608.18096v1 Announce Type: new Abstract: Assessing whether multimodal content aligns with macro-societal values, such as peace, justice, and freedom, has become an increasingly urgent challenge
arXiv:2608.18704v1 Announce Type: cross Abstract: Long-term memory is essential for agents that operate across extended interactions, yet existing memory systems and benchmarks predominantly focus on
arXiv:2510.09253v2 Announce Type: replace Abstract: Automatic systems for document understanding require multimodal models that accurately identify sensitive visual content, even in the presence of im
arXiv:2608.18532v1 Announce Type: new Abstract: Existing VLMs have achieved strong performance in video understanding, yet they struggle with long-video spatiotemporal reasoning when target objects be
arXiv:2608.18351v1 Announce Type: cross Abstract: Tool-using large language-model agents can complete a task while exercising authority that the user did not grant or the task does not need, causing e
arXiv:2608.18504v1 Announce Type: new Abstract: Universal multimodal retrieval aims to support diverse instruction-aware retrieval tasks, demanding both efficient corpus-scale matching and fine-graine
arXiv:2601.02945v2 Announce Type: replace Abstract: With the rapid development of e-commerce and digital fashion, image-based virtual try-on (VTON) has attracted increasing attention. However, existin
arXiv:2608.17567v1 Announce Type: cross Abstract: Pretrained molecular language models are increasingly used as molecular encoders for learning structure-property relationships. However, their practic
arXiv:2608.17516v1 Announce Type: new Abstract: Gender bias or other social biases in large language models (LLMs) are frequently evaluated with question answering or survey benchmarks where the LLM n
arXiv:2505.22104v2 Announce Type: replace Abstract: Shielding has emerged as a promising approach for ensuring safety of AI-controlled autonomous systems. The algorithmic goal is to compute a shield,
arXiv:2608.17247v1 Announce Type: new Abstract: Personalized agents must decide whether retrieved user memory should be used, ignored, updated, or queried before it affects a current task. We use this
arXiv:2608.17075v1 Announce Type: cross Abstract: Next-encounter ICD forecasting predicts which standardized diagnosis codes will be documented at a future visit from the longitudinal record available
arXiv:2608.17607v1 Announce Type: new Abstract: Whole-slide pathology reasoning requires models to integrate gigapixel-scale visual evidence across complete case-linked slides, yet current question-an
arXiv:2608.16931v1 Announce Type: new Abstract: Electroencephalography (EEG) foundation models have recently emerged as a promising paradigm for EEG decoding by learning reusable representations from
arXiv:2608.17223v1 Announce Type: new Abstract: Financial-news direction prediction has become a popular NLP benchmark, yet reported gains depend critically on whether the train-test split is chronolo
arXiv:2608.14711v1 Announce Type: new Abstract: AI coding agent benchmarks rank agents with the Chen et al. (2021) pass@k estimator, but current implementations misapply it: they set n to the number o
arXiv:2608.14653v1 Announce Type: cross Abstract: Prediction uncertainty is a widely adopted metric for quantifying model confidence, with downstream applications spanning model explanation, data sele