Tool use, function calling, orchestration and the protocols that let models act rather than only answer.
Video-language models and video agents can produce hallucinations that conflict with spatiotemporal evidence. Existing benchmarks mainly evaluate model hallucinations, and heterogeneous mechanisms mak...
With hate speech being ubiquitous online, automatic detection is crucial, in particular when it comes to criminally relevant social media posts. We study a variety of retrieval-based in-context learni...
Asynchronous reinforcement learning has become the standard way to scale training for language models, but the resulting policy lag biases the critic toward the stale behavior policy. Existing work on...
Long-term language-model agents rely on external memory across interactions. Atomic memories are particularly useful: their fine-grained semantic boundaries enable precise retrieval and direct compari...
Procedural memory lets language agents reuse successful routines, but reuse presumes that a stored routine remains applicable. We study what happens when that presumption is deliberately violated. The...
As large language models are increasingly deployed as tool-augmented legal agents, they introduce agentic hallucinations where tool-call and reasoning errors cascade into fabricated holdings and misci...
As Digital Twin (DT) systems evolve beyond state synchronization toward task-oriented and knowledge-driven operation, Cognitive Digital Twins (CDTs) have emerged as an extension that incorporates cogn...
Sixth-generation (6G) wireless networks are expected to provide a key infrastructure for large-scale embodied intelligence, where heterogeneous robots collaborate through low-latency connectivity, edg...
On the Navier–Stokes Millennium Prize Problem introduces an impressive result from OpenAI, who used an unreleased model to produce a resolution to the Navier–Stokes existence and smoothness problem, o...
Simon Willison reports that OpenAI used an unreleased model to produce a claimed resolution to the Navier–Stokes existence and smoothness problem, one of the seven Millennium Prize Problems. The claim has been met with controversy, as NYU professor Tristan Buckmaster and mathematician Levent Alpöge—who had been working on related problems with Claude and Codex—accused OpenAI of using their unpublished work. OpenAI has denied accessing their data and has offered to wait for Buckmaster’s publication, but will not include Alpöge as a co‑author due to a competitive relationship with his employer.
Simon Willison announces the release of ChatGPT Images 2.5, noting that OpenAI’s image generation models have processed over 3 billion images across ChatGPT Images and the GPT‑Image API. The new version improves instruction‑following across multiple turns, speeds up responses, and better preserves subjects from reference photos. Two new API model IDs—gpt‑image‑2.5‑sunburst and gpt‑image‑2.5‑flare—are available, with Sunburst recommended for precision editing and Flare for fast, high‑quality everyday generation. Willison has updated his openai_image.py CLI tool to accept reference images, demonstrating its use with a raccoon scientist prompt.
Choosing a recovery process for scale-up requires connecting laboratory results with product requirements, process costs, and scale effects. We analyze records from Pacific Northwest National Laborato...
The paper introduces the Procedural Graph, a framework that structures procedural knowledge into (procedure, relation, procedure) triplets to guide large language model agents in planning and tool usage. At each decision point, a guidance model uses the local subgraph to bias the agent’s next action, while an LLM refiner self‑evolves the graph by editing its topology based on successful versus failed trajectories. Experiments across datasets and LLMs show that Procedural Graphs consistently outperform memory‑based baselines, and the self‑evolution mechanism further improves performance without manual engineering.
DXPR is a depth‑based cross‑modal place recognition framework that matches monocular camera queries to a LiDAR map using a single vision foundation model backbone. By converting both modalities into a unified depth image representation, DXPR learns modality‑invariant global descriptors without modality‑specific encoders. A geometry‑aware overlap miner refines pairwise metric learning by computing pixel‑level overlap scores, and extensive tests on KITTI and Boreas show strong performance across seasons, weather, and day/night conditions, outperforming prior CMPR baselines.
Q2D-Web is a new large‑scale benchmark for agentic Retrieval‑Augmented Generation (RAG) systems, featuring a 190 million‑document web corpus and 70 k machine‑reformulated search queries in ten languages. It supplies three sets of relevance judgments—agent citations, production rankings, and a combined set enriched with LLM‑based labels—to evaluate first‑stage retrievers. Experiments on 13 retrievers show consistent ranking across judgment sets but significant variation across domains, languages, and query types, and demonstrate that a carefully sampled sub‑corpus can approximate full‑corpus evaluation with minimal loss in Recall@1000.
Hi-FLoop introduces a hierarchical state‑feedback framework for multi‑agent traffic simulation that reconciles decision time scales over an 8‑second rollout. The model uses eight scene‑level Worlds to maintain joint hypotheses, with an 8‑second Goal, 2‑second Preview, and 1‑second Control hierarchy, and commits only executed prefixes every 0.5 seconds to preserve factual consistency. A joint preview interaction graph and a prefix‑frozen A‑to‑B cascade enable sparse interaction refinement and accurate state recovery, achieving an overall score of 0.689987 on the H‑D public‑validation split and strong oracle‑minADE performance. whyItMatters":"The paper presents a novel multi‑timescale approach that improves consistency and realism in long‑horizon traffic simulations, as evidenced by its competitive evaluation metrics."
GoAnt is a quality‑diversity multi‑agent search framework designed to discover alpha factors from market microstructure data. It employs non‑communicating Explorer, Exploiter, and Connector workers that share an adaptive Mental Map and a Queen dispatcher to allocate evaluation budgets efficiently. On real A‑share data from 2023‑2026, GoAnt achieves quality‑weighted yields of 41.8 and 47.6 in price‑volume and order‑book settings, outperforming the best baseline by 57% and 97% under matched budgets.
Large Language Model (LLM) agents are evolving from single-session tools toward long-term personal assistants that must adapt to individual users across tasks, contexts, and interactions. This shift m...
The paper introduces a three-tier persona vector to generate diverse, realistic user inputs for evaluating tool-augmented LLM agents. The vector includes 23 dimensions: categorical demographics, continuous behavioral traits, and continuous emotional states, plus a query-complexity overlay. Experiments on 64,698 conversations show that these persona dimensions produce measurable differences in agent performance and realistic scenario-reactive behavior.
We study distributed one-dimensional mean estimation under a 1-bit communication constraint. Each agent observes one sample, drawn independently from an unknown distribution, and returns a single bit in response to a query $Q: \mathbb{R}\to\{0,1\}$ chosen by a central learner.