Am I just hallucinating
Or is there any reason why I feel like model output quality seems to be better when I use higher micro-batch values (ub) in llama-cpp? I don't really have any hard numbers or anything (just running th
Knowledge catalogue
Or is there any reason why I feel like model output quality seems to be better when I use higher micro-batch values (ub) in llama-cpp? I don't really have any hard numbers or anything (just running th
arXiv:2603.08652v2 Announce Type: replace Abstract: Recent advancements in Unified Multimodal Models (UMMs) have significantly advanced text-to-image (T2I) generation, particularly through the integra
arXiv:2608.06022v1 Announce Type: new Abstract: Epitopes determine where antibodies bind antigens and shape downstream therapeutic properties such as functional blockade and escape resistance, making
In the era of base LLM scaling (2022-2024), I believed the LLM line of research would reach a capability plateau (as later seen with base LLMs). In late 2024, after the o3 test-time compute demo, I ch
This might be a good time to mention my recent keynotes titled: 'Agentic AI is Neurosymbolic AI' I would have assumed it was fairly obvious, but in case it's not: a million-line codebase (also known a
arXiv:2608.05545v1 Announce Type: cross Abstract: Generative AI used as a capable servant has greatly accelerated intellectual work, but it also risks eroding human epistemic agency by encouraging unc
arXiv:2608.04670v1 Announce Type: cross Abstract: Large Language Models (LLMs) have transformed computational linguistics and achieved remarkable performance across numerous natural language processin
arXiv:2608.05095v1 Announce Type: new Abstract: Agents for long term reasoning require a memory that can be efficiently and effectively updated over time, as new facts and external feedback continue t
arXiv:2608.04610v1 Announce Type: new Abstract: 3D vision-language models (3D VLMs) enable spatial reasoning over multi-view scenes but suffer from substantial token redundancy due to duplicated obser
arXiv:2608.04761v1 Announce Type: cross Abstract: Self-improving agents accumulate reusable insights from prior trajectories, making retrieval increasingly important for turning accumulated experience
arXiv:2608.04071v1 Announce Type: new Abstract: Automatically generating professional multimodal reports comprising both textual analysis and visual charts from structured tabular data is a critical c
arXiv:2608.04962v1 Announce Type: cross Abstract: Reinforcement learning (RL) post-training improves the reasoning capabilities of large language models, but autoregressive rollout generation remains
arXiv:2608.04355v1 Announce Type: new Abstract: Accuracy changes after language-model self-revision are usually interpreted as changes in reasoning. We show this can fail at the answer-extraction boun
arXiv:2608.04436v1 Announce Type: new Abstract: Text-to-image (T2I) models can produce visually compelling images, yet they remain limited on open-world tasks that require complex semantic understandi
arXiv:2607.15314v2 Announce Type: replace Abstract: Healthcare AI agents handle patient consultation, clinical reasoning over text and images, interactive diagnosis, and electronic health record (EHR)
arXiv:2608.03388v1 Announce Type: new Abstract: Abductive reasoning requires forming hypotheses that explain observed evidence and revising them as new evidence becomes available. While large language
arXiv:2608.03624v1 Announce Type: new Abstract: Looped transformers have emerged as a parameter-efficient alternative to scaling depth for strong reasoning. By reusing one stack of layers across T ite
arXiv:2608.03225v1 Announce Type: new Abstract: Human-interpretable computer-aided diagnosis is crucial for clinical decision making. Concept-based models excel by providing transparent reasoning and
arXiv:2608.02980v1 Announce Type: new Abstract: Large Multimodal Models (LMMs) have achieved remarkable success on images and short videos, yet scaling them to long videos remains challenging due to f
Questions & Responses(in BOLD) below. Favorite question(s) moved to end of the thread with combined responses(removed duplicates). Be optimistic folks. I'm sure we're getting other models too apart fr
arXiv:2608.03327v1 Announce Type: new Abstract: Hybrid computer-use agents can act through screenshots or call text tools. We find that having a tool available does not settle which way the effect goe
arXiv:2608.03062v1 Announce Type: new Abstract: LLM-based CAD agents produce executable parametric programs, but their correction loops may lose evidence about satisfied requirements, faulty operation
arXiv:2608.03918v1 Announce Type: cross Abstract: Efficient long-video understanding requires vision--language models (VLMs) to reason over a small number of frames selected as sparse visual evidence.
chat : add new template for DeepSeek V4 Flash 0731 (#26398) common/chat: update DeepSeek V4 templates Align the DeepSeek V4 templates with the official encoders while keeping parser behavior out of th
arXiv:2608.00207v1 Announce Type: new Abstract: Large Language Models (LLMs) perform strongly in English medical tasks but degrade substantially in Arabic, a gap widely attributed to limited training
arXiv:2608.00355v1 Announce Type: new Abstract: Progress in large language models is often summarized using a single scalar measure, such as a time horizon, a latent ability estimate, or an aggregate
arXiv:2608.01913v1 Announce Type: cross Abstract: Deep search agents answer difficult information-seeking questions by iteratively issuing search queries to gather supporting evidence, but it remains
arXiv:2608.01359v1 Announce Type: new Abstract: Outcome-based reinforcement learning enables search-augmented language agents to learn from verifiable final answers, but its trajectory-level credit ca
arXiv:2608.01528v1 Announce Type: new Abstract: Vector symbolic architectures (VSA) are widely used for reasoning in neuro-symbolic (NeSy) AI, yet high-dimensional codebooks often create severe memory
arXiv:2608.02470v1 Announce Type: new Abstract: Vision-language models (VLMs) are increasingly deployed as reasoning agents in real-world visual assessment pipelines, yet their spatial grounding remai
arXiv:2608.01012v1 Announce Type: new Abstract: Uncommon and off-guideline cases are difficult for clinical decision support, because physicians must make a series of management decisions under diagno
Picking the right agent harness is now a crucial skill for any AI engineer. Imagine using the same model, same task, and same prompt. Now move it between two agent harnesses and the cost per success c
arXiv:2608.01930v1 Announce Type: new Abstract: Vision-language models (VLMs) are expected to revise their reasoning when visual evidence changes. Failures to do so are often attributed to insufficien
arXiv:2511.21692v2 Announce Type: replace Abstract: We investigate how well large language models (LLMs) generalize across different task difficulties, a key question for effective data curation and e
Routing for long-horizon coding agents is a big deal. @notdiamond_ai just announced a model router that works natively with Claude Code. This is huge. It picks the model and reasoning effort before ea
arXiv:2507.02778v3 Announce Type: replace Abstract: Although large language models (LLMs) have transformed AI, they still make errors and follow unproductive reasoning paths. Self-correction is vital
arXiv:2608.02217v1 Announce Type: new Abstract: Agentic multimodal reasoning extends passive image understanding by allowing VLMs to actively acquire and refine visual evidence through visual tool int
arXiv:2608.00036v1 Announce Type: new Abstract: Real-world document tasks often ask professionals to answer questions from annual reports, regulations, clinical guidelines, and technical manuals that
AI9Stars has released G9v3-39A5B an open weights language model designed to deliver even stronger reasoning capabilities than ai9stars/G9v3-3B with its 39B and 5 active experts. It is released under t
arXiv:2607.28788v1 Announce Type: new Abstract: Clinical diagnosis at hospital admission must be made rapidly from limited, incomplete evidence. Existing diagnosis-prediction benchmarks are poorly sui
arXiv:2607.29484v1 Announce Type: new Abstract: Interventional data is widely regarded as the gold standard for teaching models causal reasoning. We test this assumption in a fully controlled syntheti
arXiv:2607.28751v1 Announce Type: new Abstract: Universal multimodal embedding (UME) maps heterogeneous multimodal inputs into a shared embedding space. Existing UME models either form embeddings thro
arXiv:2607.28955v1 Announce Type: cross Abstract: AI-generated videos are becoming increasingly realistic and difficult to distinguish from authentic ones, which facilitates malicious misuse and poses
I decided to test a few questions against DeepSeek-V4-Flash-0731. Locally, I was running Unsloth's UD-Q2_K_XL quant. After I saw the surprising shape of the results, I tested against DeepSeek's offici
TL;DR: I spent 9 days developing a new quantization method for MLX models and measured 18 variants against each other on a single M5 Max MacBook Pro (128 GB). The result is the best-measuring MLX quan
arXiv:2607.27726v1 Announce Type: cross Abstract: Deep research over data lakes requires an LLM agent to investigate evidence across thousands of heterogeneous tables and passages to synthesize a repo
arXiv:2607.27595v1 Announce Type: new Abstract: Computational approaches to intertextuality have advanced from string matching to neural retrieval, yet their outputs, similarity scores and parallel-pa
arXiv:2607.26393v1 Announce Type: new Abstract: Social deduction games (SDGs) such as Werewolf have become challenging testbeds for AI agents. These games require complex social skills such as reasoni
arXiv:2603.12055v3 Announce Type: replace Abstract: Continual learning of pretrained vision-language models (VLMs) is prone to catastrophic forgetting, yet current approaches adapt to new tasks withou
arXiv:2607.27779v1 Announce Type: new Abstract: Large chest radiography archives are difficult to search because most studies are paired only with free-text reports rather than structured clinical ann
arXiv:2607.28082v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated strong capabilities in structured query generation, making them a natural choice for Text-to-SPARQL, whic
arXiv:2607.28397v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) over knowledge graphs requires retrievers that can effectively capture both graph structure and semantic informat
arXiv:2607.28451v1 Announce Type: new Abstract: Autonomous systems inevitably age, yet their artificial intelligence typically assumes hardware remains in its original condition. Batteries degrade, se
arXiv:2607.27895v1 Announce Type: cross Abstract: Mental health understanding in long-form videos requires nuanced reasoning over observable behavior, interpersonal context, and latent psychological s
arXiv:2607.28545v1 Announce Type: new Abstract: Large language models can write, patch, and search code, but oncall root cause analysis (RCA) demands something different: reasoning over noisy metrics,
arXiv:2607.28609v1 Announce Type: cross Abstract: Computer-using agents (CUAs) are advancing rapidly across the digital world. A CUA trajectory records the agent's actions, states, and reasoning. Veri
arXiv:2607.27378v1 Announce Type: new Abstract: Accurate evaluation of multimodal large language models (MLLMs) in dental panoramic radiography (orthopantomogram, OPG) is limited by the lack of fine-g
arXiv:2607.27699v1 Announce Type: new Abstract: We propose RefineSVG, a single-step closed-loop visual feedback framework that enables multimodal large language models (MLLMs) to perform high-fidelity
First of all, I want to apologize if it's off-topic or in the wrong format. Having tried Deepseek Flash with reasoning high on a conceptually difficult task, involving Machine Learning classifiers and
arXiv:2607.27380v1 Announce Type: new Abstract: Text-to-video models have achieved remarkable visual quality, yet they still struggle to generate physically consistent dynamics because the temporal ev