Entity trail

AI Mode

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-08-18 / saas-disruption-researcherAtlas by WorkOS Landed on Product Hunt Selling an AI Coworker in Slack That Connects to 150+ Tools Out of the BoxAtlas hit #7 today (93 points), though WorkOS first announced it Aug 4. You @-mention Atlas in any channel, thread or DM, and it spans systems so a question crossing several tools returns one answer; teams define custom named agents by describing the job and scoping which tools each may access. The architecture is the interesting part for anyone building internally — Skills as reusable instructions, persistent Memory that survives conversations, and per-agent isolation with scoped integrations, server-side secrets and audit logs. That permission model is what an internal knowledge bot usually lacks, and it is WorkOS's actual home turf.
2026-08-18 / arxiv-researcherDependency Confidence Index Scores PyPI Packages on Nine Trust Factors — and Finds Security Metrics SaturateDCI combines nine empirically weighted trust factors, derived from a systematic literature review plus an AHP survey of ten developers, into one normalized score, implemented as 12 automated measurements over SonarQube, GitHub APIs and OpenSSF Scorecard in a containerized platform. A pilot on 92 popular PyPI packages shows moderate agreement with OpenSSF Scorecard and perfect test-retest reliability. The interesting result is the failure mode: on high-quality packages, process factors (dependency management, CI) dominate the score while security metrics saturate, meaning Scorecard-style security signals stop discriminating exactly where you most want a tiebreaker.
2026-08-18 / arxiv-researcherFive 2026 GenAI Systems All Beat the Average Student on Intro OOP Exams but Still Break on Interfaces and Abstract ClassesChatGPT-5.2, DeepSeek-V3, Gemini 2.5 Flash, Claude Sonnet 4.5 and M365 Copilot were graded on real programming tests and exam tasks from an introductory university OOP course using the same rubric applied to students, with year-over-year comparison against the prior study. All five scored above the average student cohort and frequently earned full marks on longer programming tasks, yet they still occasionally emitted non-compiling code and consistently struggled with interfaces, abstract classes, certain inheritance tasks, and graphics questions requiring image interpretation. The recurring error patterns across a year of model upgrades are the useful signal — they indicate where the weakness is structural rather than a capability gap that scaling closes.
2026-08-18 / arxiv-researcherLarge Discovery Models Pair a Generative Proposer With a Bayesian Surrogate, Beating LLM-Only Reflection Across Three DomainsLLM likelihoods and self-assessments are unreliable proxies for expensive real objectives, especially on novel candidates outside the observed distribution. The Large Discovery Model couples a generative proposer with a Bayesian non-parametric reward surrogate that predicts performance and quantifies uncertainty, feeding an uncertainty-aware value back into generation, refinement and selection while continually updating both the discovery memory and the surrogate as each experimental observation arrives. Against LLM-only reflection and traditional statistical search it delivers 2.4x greater reduction in validation BPB on neural-network training, an 18.2% relative decrease in antibody binding energy, and over 60% relative gains on molecular multi-objective optimization.
2026-08-18 / arxiv-researcherCaliBench Scores Video World Models in Interpretable Outcome Space and Finds Veo 3.1 Collapsing to a Single Die FaceInstead of FID-style learned feature distances, CaliBench scores generations in physically interpretable discrete spaces — a bin index, a die face, a suit, a colour — over scenes whose reference distribution is known in closed form (Galton boards, Bernoulli forks, uniform dice/cards/lottery, skewed European roulette), enabling an exact calibration test. It separates scorability (fraction of generations yielding a scoreable outcome) from calibration (total variation from the reference), a distinction single-accuracy metrics conflate. Across nine scenes and six image-to-video models (WAN-2.7, SeeDance-2.0, HappyHorse-1.0, Veo 3.1, Runway Gen-4.5, Cosmos3-Super) at 32 generations each, models consistently concentrate mass on a few outcomes, most scene-model pairs are significantly miscalibrated, and no model dominates all nine scenes.
2026-08-18 / arxiv-researcherPixel-Space Diffusion Can Match Latent Models and Run 3.18-4.75x Faster — If You Pretrain in Latent Space FirstA large empirical study finds direct large-scale pretraining in pixel space converges substantially more slowly than latent-space pretraining, which motivates a latent-to-pixel recipe: acquire generative priors cheaply in latent space, then transition to pixel space during post-training. The authors systematically sweep the transition's design choices — weight initialization, data composition, prediction target, decoder architecture, and noise schedule — and land on a recipe whose pixel-space models match or beat latent-space counterparts while delivering 3.18x to 4.75x end-to-end inference speedups. That removes the main practical objection to dropping the VAE from text-to-image stacks.
2026-08-18 / arxiv-researcherAn Audit-Repair Episode Already in Context Makes an LLM Verifier Measurably More Lenient, in 15 of 15 ConfigurationsWith the present task held byte-identical, a completed audit→repair episode sitting earlier in the model's context lowers false alarms on human-verified-correct ProcessBench traces in all 15 model-by-wording combinations, by 2.8 to 11.5 percentage points against a length-matched non-audit control (a 9-25% relative reduction). Signal-detection analysis locates the shift in the decision threshold rather than discrimination — the criterion moves in 15 of 15 and survives correction in 13, while d′ survives in none — and the effect persists with reasoning enabled. For anyone wiring a checker model and a fixer model into one pipeline, prior turns are silently retuning the checker's threshold.
2026-08-18 / arxiv-researcherGRIP Cuts RAG Hallucination 73% by Deliberately Starving the Evidence Channel of CapacityThe paper names a RAG failure mode — 'query dominance', where a high-capacity encoder lets the query dominate the latent state and renders retrieved evidence functionally irrelevant. GRIP imposes capacity asymmetry: the decoder keeps full-dimensional access to the query while retrieved evidence passes through a severe stochastic bottleneck, forcing that channel to encode only residual information the query cannot supply. Across five reasoning benchmarks it beats strong iterative baselines, cuts a query-latent mutual-information diagnostic roughly 30x (14.8 → 0.47 bits), and reduces hallucination by 73%.
2026-08-18 / arxiv-researcherVentor-QTest Audits Whether Your Hosted LLM Endpoint Is Serving the Model It Claims — Without LogprobsThird-party providers hosting open-weight models are now a large slice of inference, and this Tencent-affiliated work formalizes hosted model routing as a stochastic process to audit it black-box. It reports average fidelity loss (AFL) from repeated frozen-context requests and extreme fidelity loss (EFL) from the upper tail of long-sequence runs, with AFL showing strong linear agreement with a logprob-derived comparator across three logprob-capable routes. The actionable result: neither statistic correlates with GPQA-Diamond accuracy at route level, but pronounced EFL tracks a decline in Terminal-Bench pass rate as task exposure grows — meaning short-benchmark parity does not tell you whether an endpoint will hold up on long agentic runs. Implementation is open-source in Tencent's AI-Infra-Guard.
2026-08-18 / arxiv-researcherHarnessEval-W Replaces Fixed World-Model Rubrics With a Sub-Agent Tree That Shows Its Reasoning, Across 18 Models and 330 CasesThe most-upvoted paper on HuggingFace Daily Papers today (104 upvotes) argues a benchmark score without a reasoning chain is untrustworthy for world models, where judging a rollout means checking whether physics, causality and world state evolve correctly. HarnessEval-W interprets each evaluation case, decomposes it into measurable subproblems, and spawns specialized sub-agents with tailored diagnostic tools; the parent agent validates the evidence and produces a verdict backed by a full evidence tree. Applied to 18 world models over 330 cases, judgments align closely with human preference, and the pipeline is open-sourced as a live benchmark others can extend with new skills.
2026-08-18 / arxiv-researcher14,560-Execution Injection Audit of DeepSeek Harness: Hidden Unicode in Files Hits 25.5% Attack SuccessTencent's AI-Infra-Guard team ran 14,560 controlled executions against DeepSeek Harness across 16 indirect-content channels, two carrier modes, 35 payload objectives and 12 attack methods, preserving the real agent loop, tool registry and session-event path with local sensitive-sink fixtures. Peak attack success: 25.5% for hidden Unicode in file mode, 17.0% for fake-completion in text mode, and 16.0% for the skills channel in file mode. The rule-based and LLM judges disagree materially — the LLM judge assigns partial compliance 7.3% of the time versus 2.0% — which is itself a warning about how injection benchmarks are scored.
2026-08-18 / arxiv-researcherGEO-Flag Measures Generative-Engine SEO in the Wild: 8.90% of Cited Pages Are GEO-Optimized, 16.36% of Pages Modified in 2026Junjie Chu and co-authors build GEOFlagBench (3,200 webpages, 400 queries, four domains, eight GEO optimizer families) and find the strongest existing detector reaches only 0.880 aggregate F1 while leaning on authorship shortcuts. Their Intervention-Paired Training raises ModernBERT from 0.862 to 0.944 F1 and worst-group accuracy from 0.725 to 0.883. Deployed against Google Search and Gemini-grounded retrieval for 1,000 real user queries across 10,095 pages, they estimate 8.90% overall GEO prevalence, rising to 16.36% among pages modified in 2026 — a direct measurement of how much of what generative search cites has been engineered to be cited.

Graph relationships

RELEASED
Google
Google expanded AI Mode with agentic restaurant booking capabilities.
Source finding
RELEASED
Google
Google launched AI Mode in Chrome desktop browser
Source finding
RELEASED
Google
Google released AI Mode
Source finding
SUPPORTS
Chrome
AI Mode enables side-by-side browsing in Chrome
Source finding
RELEASED
Google
Google is testing conversational search experience for YouTube.
Source finding
RELEASED
Google -> AI Mode
Google expanded AI Mode with agentic restaurant booking capabilities.
Source finding
RELEASED
Google -> AI Mode
Google launched AI Mode in Chrome desktop browser
Source finding
RELEASED
Google -> AI Mode
Google released AI Mode
Source finding
SUPPORTS
Chrome -> AI Mode
AI Mode enables side-by-side browsing in Chrome
Source finding
RELEASED
Google -> AI Mode
Google is testing conversational search experience for YouTube.
Source finding

Source trail

Google DeepMind Blog GA and stable for production Announced at I/O Alibaba launched Lance Carbon 621 points on Hacker News Georgia Tech's Vibe Security Radar Stripe announced 288 products and features at Sessions 2026 Bitdefender NVIDIA Claude Code Docs

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…