All entries

What's the best setup for Qwen3.8 27b for a 16 gig VRAM?

model-releasesr-localllama

Hello guys! I have been experimenting with qwen 3.8 for a long time and I hadn't been able to get reasonable speed. I am on a 5060Ti 16 GB, and although this gpu can game, I am aware that AI demands m

Content type

All Blog X Post Paper YouTube Reddit GitHub

When a Flatness Proxy Is Not a Function: Robustness Certificates and Training Interventions

DGX agent

arXiv:2609.38540v1 Announce Type: new Abstract: A valid curvature upper bound need not justify either a robustness certificate or an intervention on an intrinsic predictor property. We demonstrate thi

researcharxiv-cs-lg

1 Oct 2026

When a Kindergartener Solves Calculus: Measuring Capability Leakage in Role-Prompted Reasoning Models

arXiv:2609.39846v1 Announce Type: cross Abstract: We investigate the problem of role-capability leakage (RCL), in which a role-prompted reasoning model generates convincing in-role text while continui

When Can Text Replace Vision? Structural Bottlenecks in Diagram Reasoning

arXiv:2609.39142v1 Announce Type: new Abstract: Can structured text replace vision for diagram reasoning? A wrong answer after textualization can arise because the representation omits information the

researcharxiv-cs-cv

When Clipping Reverses Correction: Failure Dynamics of Pointwise Forward-KL On-Policy Self-Distillation

arXiv:2609.38995v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) trains a student on its own generated responses using feedback from the same model conditioned on privileged informat

safetyarxiv-cs-cl

When Context Changes: Understanding Update Failures in LLMs

arXiv:2609.38866v1 Announce Type: new Abstract: As preferences, goals, and facts change, LLM agents must use the current state while earlier versions remain in context. Yet they can answer with an old

Tutorials

When Correct Memory Goes Wrong: Fuzzing Persistent Memory Use in LLM Agents

arXiv:2609.38275v1 Announce Type: cross Abstract: Persistent memory helps LLM agents carry information across long interactions, but correct memory can still be used incorrectly when queries change or

tutorialsarxiv-cs-ai

When Does a Spoken Agent Have Enough Evidence to Act? The PACT-SLM Contract Test

arXiv:2609.38232v1 Announce Type: cross Abstract: Streaming spoken agents may take an external action before the available speech supports it, yet final-turn scores do not reveal whether each observed

agentsarxiv-cs-cl

When Does Randomized Oversight Align AI Agents That Can Conceal?

arXiv:2609.38262v1 Announce Type: cross Abstract: Oversight changes the evidence it relies on. We ask when randomized audits and scoring align AI agents that can conceal misconduct and alter records.

agentsarxiv-cs-ai

When Instructions Retrieve Trajectories: Diagnosing and Mitigating Generalization Failures in VLA Models

arXiv:2609.39971v1 Announce Type: cross Abstract: Vision-language-action (VLA) models can exceed 90% success on in-distribution tasks and withstand nuisance changes that preserve the required action,

safetyarxiv-cs-lg

When Integral Meets Decomposition: A Signal-Level Self-Supervised Feature Decompose Paradigm for Multi-Modal Image Fusion

arXiv:2609.39004v1 Announce Type: new Abstract: Multimodal image fusion (MMIF) aims to integrate complementary information from different modalities into a high-quality fused image and support downstr

researcharxiv-cs-cv

Hardware

When Masking Helps or Hurts Robustness in Compressed CLIP: A Pre-Deployment Diagnostic

arXiv:2609.39704v1 Announce Type: cross Abstract: This paper demonstrate that whether masking-based token pruning helps or hurts worst-group robustness can be predicted before deployment, without labe

hardwarearxiv-cs-ai

When, Not How Much: Evaluating Time-Series Foundation Models on Sparse Events

model-releasesarxiv-cs-lg

arXiv:2609.39386v1 Announce Type: new Abstract: Pretrained time-series foundation models (TSFMs) are evaluated as forecasters of future values, yet for sparse series many decisions depend only on whic

When Order Matters: First-Speaker Bias and Mitigation through Personality in Sequential Multi-Agent Debate

arXiv:2609.38964v1 Announce Type: new Abstract: Multi-agent debate (MAD) is often used to improve large language model (LLM) reasoning, but sequential debate is rarely a neutral aggregator of agents'

When Reasoning Goes Astray: Attention Dynamics of Uncontrolled Reasoning

arXiv:2609.38817v1 Announce Type: new Abstract: Large reasoning models (LRMs) improve performance on complex tasks through extended reasoning, yet the same process can degenerate into redundant verifi

When Scientific Contradictions Are Lost in Translation

arXiv:2609.38621v1 Announce Type: new Abstract: Two scientific findings can disagree without contradicting each other. Determining whether they conflict requires knowing whether they describe comparab

When the Right Answer Is Missing: An Arithmetic-Dependent Rejection Bottleneck in Jev

arXiv:2609.39496v1 Announce Type: cross Abstract: Typed decision models such as Jev offer an efficient alternative to generative LLMs in decision-making workflows by selecting directly from predefined

researcharxiv-cs-cl

Where MLLMs Fail and Why: Causal Task Decomposition for Capability Failure Diagnosis

arXiv:2609.38851v1 Announce Type: cross Abstract: End-to-end accuracy on compositional tasks records how often MLLMs fail, but cannot distinguish whether a failure reflects an intrinsic deficit in the

Where Scientific Search Agents Fail: Decision-Checkpoint Auditing of Exposure and Inspection Attempts

arXiv:2609.38670v1 Announce Type: new Abstract: Final-answer accuracy does not reveal whether a scientific-search agent failed to encounter a target paper, attempt to inspect it, or return an accepted

Where the Evidence Lives: Auditing AI Companions' Self-Descriptions

arXiv:2609.38753v1 Announce Type: cross Abstract: Companion agents describe themselves: they remember, they understand their users, the relationship has changed them. We argue that such accounts, and

agentsarxiv-cs-ai

Which Models Work Well Together? Measuring Heterogeneity for LLM Team Selection

arXiv:2609.38274v1 Announce Type: new Abstract: The performance ceiling of an LLM team is constrained not only by individual model capabilities, but also by inter-member error resonance and predictive

researcharxiv-cs-cl

Which Tasks Survive Self-Supervised Learning?

arXiv:2609.38393v1 Announce Type: cross Abstract: Same-instance self-supervised learning (SSL) learns representations by enforcing consistency across two views of the same underlying instance. This pr

Whitening Improves Robustness to Spurious Correlations in Linear Probes

arXiv:2609.39177v1 Announce Type: new Abstract: Deep neural networks tend to rely on simple features that may be spurious and thus fail to generalize. We study this problem in the setting of linear pr

researcharxiv-cs-lg

Who Owns That? Evaluating Ownership Intuitions in Large Language Models

arXiv:2609.39483v1 Announce Type: new Abstract: Ownership establishes rights over the use, control, and transfer of objects. Understanding these relations is essential for AI systems to interact appro

Who Said What, and Will It Be Remembered? Evaluating Persistent Speaker Attribution Across Meetings

arXiv:2609.39344v1 Announce Type: cross Abstract: Speech transcripts used as long-term memory must preserve both words and stable speaker identities. Existing meeting-transcription metrics either igno

Who Verifies the Graph? Misspecification Attacks on Causal Action Verification for Language Agents

arXiv:2609.40027v1 Announce Type: new Abstract: Causal action verifiers gate an agent's state-changing tool calls by checking whether each proposed intervention is identifiable against a committed act

Whose Voice Survives the Summary? A Voice-Retention Audit of LLM Employee Listening

arXiv:2609.38818v1 Announce Type: new Abstract: Organizations increasingly route employee feedback to leaders through large language model (LLM) summaries, an unaudited layer that silences already-spo

Why am I like this? (Full Chat and Image generation on a 286 Tandy 1000 TL/3)

model-releasesr-localllama

40 year tech gap? No problem! The Tandy runs DeskMind, a native DOS program. It talks over WiFi (a PicoMEM 2 card with mTCP) to a small Python server on my PC. That server drives Qwen3.8-27B (NInfer o

Tutorials

Why Do Conventional World Models Fail to Learn Cellular Automata?

arXiv:2609.39604v1 Announce Type: new Abstract: Although conventional world models - auto-regressive or diffusion models based on transformers or convolutional networks - may learn surface statistics

tutorialsarxiv-cs-ai

WinoTS: Wavelet-based Self-Distillation for Time Series Models

arXiv:2609.39337v1 Announce Type: cross Abstract: Self-supervised pre-training of time series models is currently dominated by next-token prediction and reconstruction objectives. In continuous-valued

Industry

With most information hidden, the game Stratego had stumped AI—until now

Researchers from Carnegie Mellon, MIT, NYU, and Stanford developed **Ataraxos**, a two‑neural‑network AI that defeated top human Stratego player Pim Niemeijer 15‑to‑1 (with four draws). The key innova

industryars-technica

WorkGenesis: Building the Worlds That Teach Agents to Work

arXiv:2609.39325v1 Announce Type: new Abstract: The ability of Large Language Model (LLM) agents to complete daily and professional work is receiving increasing attention. Training such agents require

Working Around the Compute Ceiling: Byte-Exact Memory in Galahad Makes LLM Reading a One-Time Cost LLM Reading a One-Time Cost

arXiv:2609.39358v1 Announce Type: cross Abstract: A transformer language model performs a bounded amount of computation per token, and recent work by Vishal Sikka, former CEO of Infosys, argues that t

World-as-Graph: Relational World Modeling Through Latent Space Graphs

arXiv:2609.38927v1 Announce Type: cross Abstract: World models aim to learn representations of real-world environments and predict their future evolution. Recent object-centric world models have made

safetyarxiv-cs-cv

WorldAuditBench: Interactive 3D World Auditing with Multimodal Agents

arXiv:2609.40325v1 Announce Type: new Abstract: As interactive 3D worlds are increasingly used to study intelligent behavior, it becomes important to develop efficient pipelines for identifying anomal

X-Planner: Event-Structured Task Planning for Embodied Intelligence

arXiv:2609.25187v2 Announce Type: replace Abstract: Task planning bridges high-level instructions and executable behavior in long-horizon manipulation, yet modern Vision-Language-Action (VLA) systems

Yggdrasil: a Layer-First 3D Scene Graph for Real-Time Querying

model-releasesarxiv-cs-ro

arXiv:2609.38640v1 Announce Type: new Abstract: Robotic agents use 3D scene graphs (3DSG) to perform tasks ranging from scene understanding to scene interaction. Although an extensive body of work add

Industry

You should be able to use your AI subscription wherever you need:

You should be able to use your AI subscription wherever you need: I am going to try and answer a bunch of questions that people have raised about Sign in with ChatGPT in the next couple of days. Befor

industrysam-altman--x

You're Hired: Strategic Model Selection for LLM Collaboration

arXiv:2609.38816v1 Announce Type: new Abstract: While multi-agent and model collaboration algorithms gain traction to combine the strengths of diverse Large Language Models (LLMs), existing systems re

agentsarxiv-cs-cl

Zero-Compute Cross-Lingual Transferability Estimation Using Typological Feature Proxies

arXiv:2609.39640v1 Announce Type: cross Abstract: Cross-lingual transfer describes how knowledge in a source language benefits a target language. Measuring it quantitatively requires broad multilingua

Zero2Repo: Can Coding Agents Build Repositories from Scratch?

arXiv:2609.38269v1 Announce Type: cross Abstract: Coding agents are increasingly asked to build software rather than patch it, yet benchmarks for from-scratch repository construction are mostly limite

Local Ai

3 models are now available on Ollama for use with /v1/systemone: • Nimble: https://ollama.com/library/nimble • Tev1 4b: https://ollama.com/l…

3 models are now available on Ollama for use with /v1/systemone: • Nimble: https://ollama.com/library/nimble • Tev1 4b: https://ollama.com/library/tev1:4b • Tev1 0.8b: https://ollama.com/library/tev1:

local-aiollama--x

A Benchmark & Dataset for Detecting AI-Manipulated Visual Evidence in the Court System

arXiv:2609.37783v1 Announce Type: cross Abstract: Photographic evidence is becoming increasingly vulnerable to forms of alteration and fabrication that existing legal and technical workflows are not w

A Character-Level Neural Approach to Sinhala Sandhi Splitting

model-releasesarxiv-cs-cl

arXiv:2609.36131v1 Announce Type: new Abstract: Sinhala Sandhi splitting recovers the constituent words or morphemes hidden inside a phonologically merged surface form. The task is important for Sinha

A Comprehensive View of Fairness through Distributional Stability

arXiv:2609.37061v1 Announce Type: cross Abstract: We view fairness as a property of distributional stability. Rather than assessing a predictor under a fixed data distribution, we study how its predic

Tutorials

A Digital Simulation Toolkit for Physics-Based Generation of Realistic Experimental Scanning Tunneling Microscopy Images

arXiv:2609.36639v1 Announce Type: new Abstract: Scanning Tunneling Microscopy (STM) is a widely used tool for characterizing surfaces of materials at the atomic scale, playing a crucial role in discov

tutorialsarxiv-cs-lg

A Dual-Track Curation-and-Classification Framework for Resolving Ground-Truth Label Noise in Operational Sentinel-2 Wheat Area Estimation

model-releasesarxiv-cs-cv

arXiv:2609.36975v1 Announce Type: new Abstract: Operational estimation of wheat-cultivated area is persistently constrained by discordance between administrative record-keeping and remotely sensed cla

A Finslerian Approach for Embedding Directed Data

arXiv:2609.37649v1 Announce Type: cross Abstract: Many datasets carry an intrinsic directionality: citations point backward in time, cells differentiate along lineages, and traffic follows preferred r

researcharxiv-cs-lg