AI News Today

Long video summarization presents significant challenges for multimodal large language models (MLLMs), particularly in maintaining temporal fidelity over extended durations and producing summaries that are both semantically and temporally grounded. We introduce LVSum, a human-annotated benchmark for evaluating long-form video summarization with fine-grained temporal alignment. LVSum comprises 72 diverse videos spanning 13 domains with an average duration of 16 minutes, each annotated with up to

Token serves as the fundamental unit of computation in modern autoregressive models, and generation length directly influences both inference cost and reasoning performance. Despite its importance, existing approaches lack fine-grained length modeling, operating primarily at the coarse-grained sequence level. In this paper, we introduce the Length Value Model (LenVM), a token-level framework that models the remaining generation length at each decoding step. By formulating length modeling as a va

Feyn Labs has released SQRL, a family of text-to-SQL models that inspect a database with read-only probes before committing to a query. The flagship SQRL-35B-A3B reports 70.6% execution accuracy on BIRD Dev, edging Claude Opus 4.6, and distills into self-hostable 4B and 9B checkpoints. The post Feyn AI Releases SQRL, a Text-to-SQL Model Family That Inspects the Database Before Writing a Query appeared first on MarkTechPost .

Perplexity's WANDR is an open benchmark and evaluation harness with 500 evidence-heavy tasks. It tests whether research agents can discover many qualifying entities and back each one with cited, re-verifiable evidence. Perplexity Search as Code leads at 0.363 soft F1 and 0.133 hard F1. The post Perplexity AI Releases WANDR: An Open Benchmark Evaluating Research Agents That Must Search Wide And Deep appeared first on MarkTechPost .

Retrieval, agents, and workflows now ship as visual and plain-English tools. This roundup covers 10 open-source no-code and low-code platforms for building LLM apps, RAG systems, and AI agents, each with its verified license, repository, and best-fit use case. The post 10 Open-Source No-Code AI Platforms for Building LLM Apps, RAG Systems, and AI Agents appeared first on MarkTechPost .

<!-- SC_OFF --><div class="md"><p>The Google DeepMind-sponsored Kaggle challenge "Measuring Progress Toward AGI - Cognitive Abilities" asked participants to design new cognitive-science-based AI benchmarks and they just announced the results this week.</p> <p>In my two posts I present evidence that deepmind & kaggle rewarded a nonsensical number generation machine and a litany of unfounded claims with 25k and a grand prize stamp.</p> <p>What the authors of the work I analyze intended to do was t

<!-- SC_OFF --><div class="md"><p>How to submit to AI alignment track? I can only see these at openReview:<br/> AAAI 2027</p> <ul> <li><a href="https://openreview.net/group?id=AAAI.org/2027/Artificial_Intelligence_for_Social_Impact_Track">AAAI 2027 Artificial Intelligence for Social Impact Track</a></li> <li><a href="https://openreview.net/group?id=AAAI.org/2027/Conference">AAAI 2027 Conference</a></li> <li><a href="https://openreview.net/group?id=AAAI.org/2027/Innovative_Applications_of_AI">AAA

<!-- SC_OFF --><div class="md"><p>I built a small web app that lets you run tabular foundation models (currently just Google's TabFM) on spreadsheets without writing any code.</p> <p>Just drop in a CSV/Excel file, click a column header to mark what to predict, hit predict. Rows where the target cell is filled become the in-context examples and empty ones get predicted, right on the grid.</p> <p>A lot of people who'd benefit from these models aren't programmers, so I wrapped it in a UI anyone can

<!-- SC_OFF --><div class="md"><p>Started testing a private qwen 35B moe capacity LLM runtime on s26 ultra, early testing shows that active model footprint can fit within the device’s memory limits.( not sharing the methods or architecture used) and results suggest roughly 90 input processing t/s achievable after optimisation and output generation is around 8 tokens/s on this mobile.</p> <p>Point is i learned ai ml based on my interest and no formal PhD , I have compute and resources to test. An

<!-- SC_OFF --><div class="md"><p>I have released a model that I have been working on for ~6 months off and on. I've been enjoying listening to spatial music, but there is a lot of music out there with no real quality spatial mix. So I decided to make a model to convert stereo to spatial.</p> <p>I started by making a flow-matching diffusion model that operated in the latent space of a separately trained VAE (<a href="https://github.com/Eps-Acoustic-Revolution-Lab/EAR_VAE">EAR-VAE</a>). I used th

Lowest cost per token from extreme codesign maximizes intelligence per dollar for post-training in the agentic era.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. There’s a lot of hype around perimenopause. Don’t buy it. Perimenopause used to be considered taboo, but not anymore. Thanks at least in part to TV doctors and social media influencers,…

Perimenopause has entered the chat. Perimenopause—and its better-known relative, menopause—used to be considered taboo. Not anymore, thanks at least in part to TV doctors and social media influencers. Perhaps it’s my age, but these days, both my algorithm and my conversations with friends increasingly swing toward perimenopause. Menopause is defined as the life stage that…

arXiv:2607.13918v1 Announce Type: cross Abstract: Serial verification gates are a core reliability primitive in LLM harnesses: a candidate answer is returned only if $k$ verifier calls all accept it. Under conditionally independent gates, the recent Odds Law (arXiv:2606.15712) shows that posterior log-odds grow linearly in $k$, so failure decays exponentially, and states that "a tight theory of partially correlated verifier cascades remains open." This note gives a minimal such theory. Modeling

arXiv:2607.13921v1 Announce Type: cross Abstract: Languages with rich static semantics, such as Rust, provide stronger guarantees for AI-generated code, but their strictness makes generation more difficult. Off-the-shelf compilers can provide useful feedback post-generation, but does not guide intermediate generation steps, such as those during autoregressive LLM decoding. Constrained decoding intervenes earlier by rejecting invalid tokens during sampling, but requires white-box model access and

arXiv:2607.13978v1 Announce Type: cross Abstract: Music-driven dance generation aims to produce human motion that is both rhythmically synchronized and semantically consistent with music. While recent neural approaches have achieved impressive visual realism, they typically model motion as a continuous signal and neglect its compositional nature, making generated dances structurally incoherent and difficult to control. In this work, we introduce a structure-aware framework that models choreograp

arXiv:2607.13998v1 Announce Type: cross Abstract: The rapid proliferation of Agentic Artificial Intelligence fundamentally disrupts traditional customer loyalty paradigms. As AI evolves from passive recommendation algorithms to autonomous, goal-directed agents capable of executing purchasing decisions, the conventional understanding of consumer-brand relationships requires a structural reevaluation. By synthesizing extant literature across human-machine teaming, consumer decision-making, and alg

arXiv:2607.14006v1 Announce Type: cross Abstract: Penetration testing traditionally evaluates whether adversaries can exploit weaknesses in software, infrastructure, configurations, or operational controls to achieve security-relevant compromise. This paradigm remains necessary for AI-enabled systems, but it is no longer sufficient. In such systems, adversaries may influence prompts, retrieved content, sensor inputs, training data, memory, tools, or human-AI interaction loops to alter system beh

arXiv:2607.14018v1 Announce Type: cross Abstract: We investigate how each component of the Transformer feedforward block architecture design determines how much rank survives across depth at initialization. We reinterpret skip connections and normalization, long understood as controlling magnitude, as mechanisms for preserving gradient rank across depth, since the very matrix multiplications and nonlinear activations that make the network expressive also reduce the rank. We show that skip connec

arXiv:2607.14037v2 Announce Type: cross Abstract: Agentic coding tools are increasingly capable of generating and submitting pull requests (PRs) to software projects, introducing new forms of human-agent collaboration in software development. While prior studies have examined PR-level outcomes of agent-generated contributions, less is known about how agentic coding tools are adopted and managed at the project level. In this paper, we analyze 25,264 agentic PRs from 2,361 popular GitHub repositor

arXiv:2607.14041v1 Announce Type: cross Abstract: Historical Manchu OCR must accommodate various visually distinct writing styles, including regular script, running script, and the semi-cursive chancery hand used in palace memorials, despite limited labeled data. We study a multi-expert system that reuses checkpoints from an iterative fine-tuning process as domain specialists and uses a lightweight page-level image classifier to dispatch pages by visual style. When the checkpoint pool lacks a su

arXiv:2507.19593v3 Announce Type: replace Abstract: Classical game-theoretic models typically assume rational agents, complete information, and common knowledge of payoffs - assumptions that are often violated in real-world MAS characterized by uncertainty, misaligned perceptions, and nested beliefs. To overcome these limitations, researchers have proposed extensions that incorporate models of cognitive constraints, subjective beliefs, and heterogeneous reasoning. Among these, hypergame theory e

arXiv:2510.10002v3 Announce Type: replace Abstract: As large language models (LLMs) are increasingly deployed in sensitive everyday contexts -- offering personal advice, mental health support, and moral guidance -- understanding their behavior in navigating complex moral reasoning is essential. Most evaluations study this sociotechnical alignment through single-turn prompts, but it is unclear if these findings extend to multi-turn settings, and even less clear how they depend on the interaction

arXiv:2601.20379v2 Announce Type: replace Abstract: Large language models (LLMs) struggle with complex, long-horizon reasoning due to instability caused by their frozen policy assumption. Current test-time scaling methods treat execution feedback merely as an external signal for filtering or rewriting trajectories, without internalizing it to improve the underlying reasoning strategy. Inspired by Popper's epistemology of "conjectures and refutations," we argue that intelligence requires real-tim

arXiv:2602.11619v2 Announce Type: replace Abstract: Running the same LLM agent on identical inputs yields 2.3-4.2 distinct action sequences per 10 runs; this behavioral variance constitutes a training-free, black-box uncertainty signal that instantiates selective classification and distribution-free calibration for agentic systems. Across 8,000 runs of four models on 200 HotpotQA questions, consistent tasks (at most 2 unique paths) achieve 82-87% accuracy while inconsistent tasks (4 or more path

arXiv:2603.00546v2 Announce Type: replace Abstract: Using Multimodal Large Language Models (MLLMs) as judges to achieve precise and consistent evaluations has gradually become an emerging paradigm across various domains. Evaluating the capability and reliability of MLLM-as-a-judge systems is therefore essential for ensuring trustworthy assessment. Existing judge benchmarks categorize samples by task types but fail to capture the fundamental judgment capabilities required for reliable evaluation.

arXiv:2603.16307v3 Announce Type: replace Abstract: Remote sensing underpins crucial applications such as disaster relief and ecological field surveys, where systems must understand complex scenes and constraints and make reliable decisions. Current remote-sensing benchmarks mainly focus on evaluating perception and reasoning capabilities of multimodal large language models (MLLMs). They fail to assess planning capability, stemming either from the difficulty of curating and validating planning t

arXiv:2605.07323v2 Announce Type: replace Abstract: Discovering governing differential equations from observational data is a fundamental challenge in scientific machine learning. Existing symbolic regression approaches rely primarily on quantitative metrics; however, real-world differential equation modeling also requires incorporating domain knowledge to ensure physical plausibility. To address this gap, we propose DoLQ, a method for discovering ordinary differential equations with LLM-based q

arXiv:2606.06223v2 Announce Type: replace Abstract: Language-model agents act through repeated cycles of observation, reasoning, and action selection, making safety monitoring depend on both internal model state and environment context. We study reward-hacking monitors in ReAct-style agents acting in Gameable ALFWorld and WebShop. Agents are instrumented with activation-based reward-hack scores, token-level entropy, and decision-context features. We find that adapters fine-tuned on \textit{Schoo

arXiv:2606.16944v2 Announce Type: replace Abstract: Theory of mind (ToM), the capacity to ascribe mental states to others and use those ascriptions for prediction and inference, is widely assumed to be essential for effective human-machine integration. Existing AI-ToM models address \emph{how} to mentalize, but leave the question of when largely unaddressed. The central question is: under what situational and agent-level conditions is ToM engagement causally warranted in conflict? This paper pre

arXiv:2607.14194v1 Announce Type: new Abstract: Text-to-video (T2V) generators can synthesize realistic and temporally coherent videos, but controllably removing a target concept from a generator remains difficult. Unlike text-to-image concept erasure, T2V unlearning must suppress a target concept that may persist across frames while preserving non-target subjects, actions, scenes, and temporal structure. We propose \textbf{SIRUS}, a training-free inference-time framework for concept-level T2V u

arXiv:2607.14228v1 Announce Type: new Abstract: In this paper, we ask whether vision foundation models construct representations that reflect the intrinsic properties of 3D Euclidean space. Unlike previous works that probe 3D awareness of vision features by regressing image-centric quantities such as depth or normals, we investigate the relation between the structure of the space of visual features and the group of Euclidean transformations $SE(3)$. We propose a set of probes to evaluate this re

arXiv:2607.14248v1 Announce Type: new Abstract: Lane boundary detection is a critical component in autonomous driving systems and has been rigorously studied in regular driving scenarios. However, it is less explored in vehicle racing, where the car moves at higher speeds across more extreme road geometries. To study this problem, we introduce a new dataset for 3D lane detection in racing, featuring >$250$k images from multiple camera feeds and inertial measurements taken with a Lexus LC 500 dri

arXiv:2607.14264v1 Announce Type: new Abstract: Automated chest CT report generation remains challenging because clinically faithful reporting requires both whole-volume understanding and accurate description of localized anatomical findings. Here we developed and retrospectively evaluated MonteRET, a region-aware retrieval-enhanced framework for generating chest CT findings sections. MonteRET integrates global CT features with region-level anatomical representations, retrieves clinically releva

arXiv:2607.14287v1 Announce Type: new Abstract: Defect segmentation in additive manufacturing (AM) X-ray computed tomography (XCT) images remains challenging due to severe class imbalance and large distribution shifts across scan conditions. Although recent foundation models such as the Segment Anything Model (SAM) provide strong general-purpose segmentation priors, their natural-image pre-training transfers poorly to the AM XCT domain, where defects appear as subtle non-semantic microstructural

arXiv:2607.14305v1 Announce Type: new Abstract: In this paper we propose DCVC-Mamba (DCVC-MB), a neural video codec framework for B-frame coding. Our approach incorporates an IBP frame strategy for low-delay B-frame coding, a spatio-temporal fusion model based on state-space models for bidirectional temporal prediction, and an entropy-aware skipping mechanism that selectively omits coding certain latents to reduce entropy coding times. In addition to our model contributions we also implement two

arXiv:2607.14333v1 Announce Type: new Abstract: Vision Language Models (VLMs) demonstrate strong perceptual abilities but remain limited in tasks requiring analytical reasoning across multiple visual states, such as multi-image comparison, change detection, and multi-step visual inference. These capabilities are critical for real-world multimodal applications where reasoning must be grounded in systematic differences between visual contexts. However, existing benchmarks rarely require both expli

arXiv:2607.14415v1 Announce Type: new Abstract: Computed Tomography (CT) carries significant ionizing radiation risks, driving the need for sparse-view reconstruction. Implicit scene representations (ISRs) address this by recovering continuous volumetric attenuation fields directly from sparse projections, and recent geometry-aware extensions jointly model surface geometry alongside attenuation to improve fidelity and enable clean tissue segmentation without manual thresholding. However, these m

arXiv:2607.14423v1 Announce Type: new Abstract: Frozen self-supervised vision models can align parts of generic objects, but it remains unclear whether this correspondence extends to human faces, where global layout is shared while identity-specific appearance varies sharply. We test whether frozen DINOv3 features define a region-level facial coordinate system: a feature space in which eyes, brows, nose, mouth, skin, and hair remain distinguishable across people and across time without face-spec

arXiv:2607.14470v1 Announce Type: new Abstract: Few-view surface reconstruction recovers the visible surfaces of a scene from a few posed RGB images, providing the 3D models that robots need to explore and interact online. On mobile platforms, the reconstruction must be fast and geometrically accurate while keeping a small memory footprint to ensure safe and efficient operation. 3D Gaussian Splatting (3DGS) offers a high-fidelity scene representation, but building it from a few views is ill-pose