Knowledge catalogue
I tested Jev and five LLM judges across ten Arize Phoenix evaluators to measure how often their answers changed alongside accuracy, cost, and latency. The post What Jev’s probabilities reveal that rep
Every week, I talk with founders who are building at an unbelievable pace. Teams are moving from inception to product-market fit faster than ever, with foundation models wired deeply into their core p
Cool to see classification models resurge in the era of agents. The effects of Jev are insane. Also very cool to see our emotion dataset built at @dair_ai used to test this new model, Julia-1. I spent
Hardware RTX 5090 (32 GB) + RTX 4070 Ti Super (16 GB) = 48 GB VRAM 32 GB DDR5 6200 RAM Arch Linux, KDE on the 5090 (takes ~1-1.5 GB VRAM) daily driver OrcaRouter Qwen3.8-27B Uncensored, block-FP8 (28.
Finally got 3.8-Flash-Next running on my M4Pro 48GB Mac with https://huggingface.co/ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF Dense 3.8-27B is just faster... and maybe better due to quantization lev
Upgraded from 3x RTX 3090s to 2x RTX 5090s on my homelab server and picked up a solid speed jump on top of it from a software update (speculative decoding + NVFP4). Setup: llama.cpp (build b11216), ru
arXiv:2609.29204v1 Announce Type: new Abstract: Vision-language-action (VLA) models offer strong local control and instruction following but often struggle with long-horizon tasks requiring persistent
arXiv:2609.29212v1 Announce Type: new Abstract: Large language models can decompose mobile-manipulation goals into long action sequences, but the resulting plans remain reliable only while their world
arXiv:2609.28972v1 Announce Type: cross Abstract: Cross-country recommendation on modern e-commerce platforms is typically deployed with disjoint user and item ID spaces across markets, removing the s
arXiv:2609.29556v1 Announce Type: new Abstract: Multimodal emotion recognition is a key research area in affective computing, with applications in sentiment analysis, intelligent customer service, and
arXiv:2609.29043v1 Announce Type: cross Abstract: General Purpose Service Robot (GPSR) tasks, as defined in the RoboCup@Home benchmark, require robots to interpret diverse natural language commands an
arXiv:2609.28796v1 Announce Type: cross Abstract: Current automated gait analysis for clinical applications relies on uninterpretable black-box classifiers. Although Vision-Language Models (VLMs) offe
arXiv:2609.29613v1 Announce Type: cross Abstract: We present a modular and multi-level framework for the differential diagnosis of malignant melanoma. Our framework integrates contextual information a
arXiv:2609.29812v1 Announce Type: new Abstract: Looped Transformers have attracted substantial attention as a parameter-efficient approach to increasing computational depth through repeated applicatio
arXiv:2609.28487v1 Announce Type: new Abstract: News headlines frame public issues both by what they select and by how they word it, yet computational framing work typically collapses these operations
arXiv:2609.29370v1 Announce Type: cross Abstract: Science, technology, and innovation policies are crucial for competitiveness, yet their diversity and scale make them difficult to map and monitor con
arXiv:2609.28477v1 Announce Type: cross Abstract: Claims data can show that provider behavior changed but cannot by itself explain why. FWA (fraud, waste, and abuse) review requires identifying materi
I recently started using my Mac Mini M4 as an always-on home server. It stays powered on most of the time, and right now I’m running my Discord bot on it. I’m thinking of gradually turning it into my
arXiv:2609.29167v1 Announce Type: new Abstract: Banking assistants must use account-specific information to answer requests and, in many cases, take actions through tools. Evaluating only the final re
arXiv:2609.29444v1 Announce Type: cross Abstract: Deep search requires LLM agents to decompose complex queries, search for evidence, and synthesize grounded answers, yet existing ReAct-style agents su
arXiv:2609.28585v1 Announce Type: cross Abstract: Multi-step tool-calling LLM agents rely on host runtimes to preserve state across turns. When a runtime carries an external tool return into later mod
arXiv:2609.30094v1 Announce Type: new Abstract: Large language models increasingly operate as persistent assistants in user-facing, shared-session, and tool-augmented settings. When users disclose sen
arXiv:2609.28693v1 Announce Type: new Abstract: Large Language Model (LLM) agents struggle to scale safely when exposed to vast enterprise toolsets. Providing an agent with access to every internal to
Prefill step size affects both PP performance and your drafter fetching logits for MTP (affects dflash as well, mlx-vlm needs to be patched slightly to support chunked prefill for dflash). It also nee
arXiv:2609.29837v1 Announce Type: new Abstract: We introduce PUBG Ally, an embodied agent for PUBG: BATTLEGROUNDS that can reason, act autonomously, and play alongside players as a voice-enabled teamm
I maintain a starter kit where the LLM provider is a runtime-swappable DB row, not a hardcoded constant — OpenAI, Anthropic, or a local Ollama instance. Mostly so I can demo without burning API credit
arXiv:2602.23960v2 Announce Type: replace-cross Abstract: How natural speech is represented in the brain constitutes a major challenge for cognitive neuroscience. Reconstructing the speech envelope an
arXiv:2602.11700v2 Announce Type: replace-cross Abstract: Tabular prediction can benefit from in-table rows as few-shot evidence, yet existing tabular models typically perform instance-wise inference
The era of omni agents is upon us. This is a great report by the Qwen Team on their omni-modal agents. They present Qwen3.8-Omni-Flash, a natively multimodal model trained for long-horizon agent tasks
arXiv:2609.27844v1 Announce Type: new Abstract: Claim denial management costs U.S. healthcare approximately $260 billion annually in administrative overhead. Large Language Models (LLMs) and Retrieval
vulkan: int8 coopmat1 matmul implementation for AMD RDNA3 and RDNA4 (#27952) vulkan: add int8 coopmat quantized matmul shader apply scales inline use scalar sums probe and directly access coopmat valu
arXiv:2607.22115v2 Announce Type: replace-cross Abstract: Given a database S and a natural language question Q, text-to-SQL systems aim to generate an SQL query that correctly answers Q when executed
arXiv:2603.13768v2 Announce Type: replace-cross Abstract: Despite the strong performance of large audio language models (LALMs) in various tasks, exactly how and where they integrate acoustic features
arXiv:2609.27014v1 Announce Type: cross Abstract: Legal contracts are structurally complex documents in which contradictions may emerge across distant and interconnected provisions. Although large lan
arXiv:2609.28041v1 Announce Type: new Abstract: Conference policies distinguish using Large Language Models (LLMs) to polish one's own review from delegating the critique, but current Artificial Text
arXiv:2609.27770v1 Announce Type: new Abstract: Autonomous web agents, powered by Large Language Models (LLMs), have garnered significant attention for automating various web-based tasks with multi-st
arXiv:2609.27656v1 Announce Type: cross Abstract: Physical intelligence requires more than predicting how the world may evolve: predictions must remain actionable as the world continues to change. We
arXiv:2603.28790v2 Announce Type: replace-cross Abstract: In edge computing, the stochastic and bursty nature of serverless workloads challenges autonomous resource orchestration. Traditional reactive
I'm about to buy a Mac Studio mainly for running LLMs locally and I'm stuck between two configs: M5 Ultra (30/64) with 96GB: 1.2 TB/s bandwidth, roughly 1.7x faster generation and much faster prefill
arXiv:2609.26855v1 Announce Type: cross Abstract: Relational Deep Learning (RDL) models multi-table databases as heterogeneous temporal graphs, and graph transformers currently achieve state-of-the-ar
arXiv:2609.27913v1 Announce Type: new Abstract: We study the problem of aggregating opinions from multiple black-box experts in noisy, conflict-prone settings where expert reliability varies across in
Scribd, Inc. is home to one of the world's largest collections of human-created content. Scribd’s products leverage one of the world's largest collections of human-created content and intelligent tool
arXiv:2609.28006v1 Announce Type: new Abstract: Decoder-only Transformer language models cache keys and values (KV) to reuse past computation during generation. Sharing KV across layers saves storage
arXiv:2609.26820v1 Announce Type: cross Abstract: Physiological time series such as electrocardiograms (ECG) and electroencephalograms (EEG) exhibit complex temporal structure, substantial acquisition
arXiv:2609.27340v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used as natural-language interfaces for robotic systems, yet their integration with Robot Operating System
arXiv:2609.27298v1 Announce Type: new Abstract: Long-horizon agents continuously accumulate interaction history during task execution, yet the importance of past interactions changes as the agent stat
arXiv:2609.28466v1 Announce Type: new Abstract: Advances in generative models have improved video fidelity, enabling long-horizon generation, interactive world modeling, and evolving visual environmen
arXiv:2609.27868v1 Announce Type: cross Abstract: Octree-based 3D Gaussian Splatting organizes anchors into multi-level hierarchies for level-of-detail rendering, but features at different levels are
arXiv:2609.27510v1 Announce Type: cross Abstract: Modern large language models are pretrained on massive datasets, making it difficult to prevent benchmark data from entering their training sets and u
arXiv:2609.27064v1 Announce Type: new Abstract: A joint fact-verification score assesses answers and submitted evidence together. When the score improves, how much of the gain remains if the answers a
arXiv:2609.26929v1 Announce Type: new Abstract: People hold diverse, sometimes conflicting values, so no single aligned model can satisfy everyone. Pluralistic alignment therefore calls for steerable
arXiv:2609.26781v1 Announce Type: new Abstract: A multi-agent system can reduce latency on complex tasks by executing work concurrently. Several pioneering harness frameworks support multi-agent syste
arXiv:2609.25666v1 Announce Type: new Abstract: We present and tackle two problems associated with deploying Foundation Models (FMs) on Embodied Agents performing navigation: 1) Training bias in FMs l
arXiv:2410.14948v2 Announce Type: replace Abstract: Medical multimodal large language models (MLLMs) can perform well on existing medical visual question answering (MedVQA) benchmarks, but their train
arXiv:2609.23695v1 Announce Type: new Abstract: Recent advances in Physical AI have accelerated the use of foundation models in autonomous systems such as unmanned aerial vehicles (UAVs), which must p
What it is. stuntd is a local proxy for typed LLM decisions: choice, yes/no, score. Your app keeps calling its provider through it. stuntd records the decisions, trains a small head on top of Laya (th
arXiv:2609.22712v1 Announce Type: new Abstract: Agentic AI systems built on large language models can plan over multiple steps, use external tools, retain information in memory, and coordinate with ot
arXiv:2602.18307v2 Announce Type: replace-cross Abstract: Large language models have achieved striking results in interactive theorem proving, particularly in Lean. However, most benchmarks for LLM-ba
arXiv:2609.24362v1 Announce Type: new Abstract: Sandboxed computer environments support multi-step reasoning with tools, executable programs, and persistent files, yet their extension from language mo
arXiv:2609.22100v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) improves language models with retrieved evidence, but processing many long passages is costly and can introduce dis