Search: “Common Sense Media”

Meta pledges to overhaul kids’ safety protections, pay $17 billion to settle social media case

Meta settles states' kids' online safety lawsuit for $17 billion, agreeing to landmark usage limits, age restrictions, and independent auditing.

Meta agreed to pay $17 billion to settle a civil suit from nearly every US state and territory alleging it hid research showing Facebook and Instagram are addictive to minors and violated COPPA by collecting data on children under 13. The settlement imposes reforms including two-hour daily limits for users under 18, a midnight-to-6am usage block, non-personalized feed options, and an independent auditor. Meta also settled separately with Texas for about $1 billion.

The Record·Aug 26, 18:53 UTC · 21d agoPolicy & legal

Diffusion TV: Experiencing Diffusion Models through Tangible, Embodied Interaction

Researchers present Diffusion TV, a CRT-based installation where antenna manipulation lets audiences physically experience diffusion model denoising.

Diffusion TV is an interactive installation built around a modified CRT television where turning the antenna controls the clarity of AI-generated images and sounds, mirroring the denoising process of diffusion models. Three channels present AI-generated animals from the past, present, and future within a temporal and ecological narrative. The authors frame the work as an embodied, non-verbal alternative to explainable AI that highlights intermediate generative states rather than final outputs.

arXiv cs.AI / cs.LG / cs.CL·Sep 4, 17:51 UTC · 12d agoAI research

Anthropic’s Text Watermarking Proves AI Companies Do Not Care at All About Writing

Anthropic will watermark Claude outputs by biasing low-stakes word choices, a method it says complies with EU AI regulations.

Anthropic detailed how future Claude versions will carry a statistical watermark by altering the source of randomness used to pick among near-synonymous words, adding no hidden characters or metadata; a key holder can compute a probability that text was Claude-generated. The company says internal testing showed no impact on quality, creativity, or readability, and frames the change as compliance with new EU AI regulations. Critics including John Gruber and Jeff Jarvis argue the method treats synonyms as interchangeable and devalues writing, a view echoed in this 404 Media opinion piece.

404 Media·Aug 18, 15:54 UTC · 29d agoAI industry

SenseNova-U1.5: Towards Native Unified Visual Intelligence

SenseTime releases SenseNova-U1.5, an 8B-MoT encoder-free multimodal model unifying visual understanding, reasoning, and generation with native 4K resolution.

SenseNova-U1.5 is an 8B mixture-of-transformers multimodal model with an encoder-free, VAE-free architecture that understands, reasons about, and generates visual content at native resolutions up to 4K. Post-training optimizes specialized experts for visual aesthetics, bilingual text rendering, infographic generation, and image editing, consolidated through multi-expert on-policy distillation. Evaluations report gains in image fidelity, text rendering, multi-reference editing, and instruction following. The team plans to open-source training code including supervised fine-tuning, reinforcement learning, and on-policy distillation.

Hugging Face daily papers·Sep 9, 20:00 UTC · 6d agoModel release

I’ve been deepfaked: What do I do?

ESET outlines steps for deepfake victims: preserving evidence, using platform reporting tools, and legal remedies like the US TAKE IT DOWN Act and StopNCII.org.

ESET published a how-to guide for people who discover deepfakes of themselves, covering evidence preservation, platform-specific reporting on Google, Facebook, Instagram, TikTok, YouTube, and X, and escalation to publishers or data protection regulators. It notes the US TAKE IT DOWN Act criminalizes non-consensual intimate imagery (NCII) and requires 48-hour takedowns, while UK and EU laws add creation offenses and GDPR Article 17 erasure rights. Services like StopNCII.org and TakeItDown.NCMEC.org hash images so participating platforms such as Meta, TikTok, Reddit, and X can find and remove matching copies.

ESET WeLiveSecurity·Sep 2, 11:48 UTC · 14d agoAI safety & security1

Cyber Security News

Microsoft's September Patch Tuesday fixes nearly 1,000 vulnerabilities; CISA warns two of the flaws are being actively exploited.

Microsoft's September 2026 Patch Tuesday release addresses close to 1,000 vulnerabilities, with CISA warning that two of the bugs are being actively exploited. The Record's homepage digest also lists briefs including a US offer of $10 million for information on an Iranian hacker accused of attacking critical infrastructure and a leak of health data on more than 9.5 million people from the Aesto record system. Additional briefs cover a Russian suspect's extradition to the US for bank account takeovers, US-British coordination on scam center takedowns, G7 guidance on quantum cyber threats, a large DDoS attack on Norwegian public services, and Slovenian casinos reopening after a cyberattack.

The Record·Sep 9, 20:23 UTC · 6d agoVulnerability in the wild

The Smishing Deluge: China-Based Campaign Flooding Global Text Messages

Unit 42 attributes a global smishing campaign with 194,000+ phishing domains impersonating tolls, banks, and couriers to the Smishing Triad.

Palo Alto Unit 42 attributes ongoing smishing texts about toll violations and package misdelivery to the Smishing Triad, targeting U.S. residents since April 2024. Researchers identified 194,345 FQDNs across 136,933 root domains registered since January 2024, mostly via Hong Kong registrar Dominet (HK) Limited with Chinese nameservers and hosting concentrated on U.S. cloud services. The decentralized campaign impersonates banking, cryptocurrency, e-commerce, healthcare, law enforcement, and social media services, and its scale points to a large phishing-as-a-service operation. Phishing pages harvest national ID numbers such as Social Security numbers, home addresses, payment details, and login credentials.

Palo Alto Unit 42·Aug 17, 12:45 UTC · Aug 17, 2026Phishing & fraud in the wild1

The Attention Triangle in Audio-Video Models

Researchers analyze the 'attention triangle' in audio-video diffusion models, showing bias-driven cross-attention routing causes semantic leakage and proposing inference-time interventions that improve grounding.

A study probes the three cross-attention edges linking text, audio, and video streams in audio-video diffusion models. It finds the audio-video edge is bidirectional and shaped by parameter-encoded biases, so prompts in tension with learned priors can be overridden, producing visually canonical but incorrect outputs. Attention-derived signals are used as diagnostics and to guide inference-time interventions that improve cross-modal semantic grounding while preserving generation quality.

Hugging Face daily papers·Sep 2, 20:00 UTC · 13d agoAI research

Mistral raises €3B as sovereign AI becomes big business

Mistral AI raised a €3B Series D at a €21B+ valuation, led by Samsung Electronics, to scale European compute capacity and sovereign AI services.

French AI lab Mistral AI raised €3 billion (~$3.58B) at a post-money valuation above €21 billion, which it calls the largest equity round ever completed by a European technology company. Samsung Electronics led the round, with EQT's Scaleup Europe Fund and PSG Equity as co-leads; a16z, Nvidia, Salesforce Ventures, Advent, BlackRock, and Luxembourg also participated. The company plans to build 1 GW of European compute capacity by 2030, offer region-selectable query processing, and host third-party open-weight models as part of a sovereign AI strategy. Mistral operates in 20 countries and targets governments and enterprises seeking control over AI models and data residency.

TechCrunch · AI·Sep 8, 14:17 UTC · 8d agoAI industry

Enhancing Accessibility of Medical Texts through Large Language Model-Driven Plain Language Adaptation

Study shows LLMs with Mixture-of-Agents and QLoRA finetuning effectively simplify medical texts into plain language while preserving content.

The paper evaluates Plain Language Adaptation (PLA) using GPT-4o-mini, Gemini-1.5-pro, and LLaMA in zero-shot and few-shot settings. It compares prompting strategies, QLoRA finetuning across models, and integrates Mixture-of-Agents (MoA) techniques for robustness. Results demonstrate LLM-driven PLA makes healthcare texts more comprehensible while retaining essential content.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 16:33 UTC · 1d agoAI research

Simulation: the new Scaling Law — Joon Sung Park, Simile AI

Simile AI raised a $2B Series B from GreenOaks and Index Ventures to scale human-behavior simulation for Fortune 100 clients like CVS.

Simile AI, co-founded by Generative Agents researcher Joon Sung Park, announced a $2 billion Series B backed by GreenOaks and Index Ventures, with Fei-Fei Li and Andrej Karpathy among backers. The company runs tens of millions of simulations for Fortune 100 clients including CVS, reporting 85-99% accuracy versus human focus groups and digital twins of 1,000 real people at 85% behavioral accuracy. The long-term ambition is foundation models of human behavior, post-trained on interviews, transaction data, and randomized controlled trials, potentially simulating all 8 billion people.

Latent Space·Aug 21, 23:37 UTC · 25d agoAI industry

‘Own the Narrative’: Leaked Flock Guide Shows How It Teaches Cops to Promote Its Tech

404 Media·Aug 3, 17:27 UTC · Aug 3, 2026Threat actor

I Think the Military Commissary Freezers Were Hacked

Refrigeration failures at six-plus US military commissaries prompt speculation of a cyber attack on DeCA's remote monitoring systems; Pentagon acknowledges possible disruption.

The author documents near-simultaneous freezer and refrigeration failures at confirmed installations including Fort Huachuca, F.E. Warren AFB, Fort Irwin and Travis AFB on August 26-27, with freezers entering defrost mode that heated and spoiled food. DeCA's Remote Monitoring Control System controls defrost across roughly 182 locations, and an unverified comment attributed the Fort Huachuca failure to a network issue. Stars and Stripes and Military Times independently reported the multi-base failures, and the Pentagon acknowledged a 'possible refrigeration disruption,' though no evidence of hacking has been confirmed.

Lobsters · security·Sep 3, 03:20 UTC · 13d agoData breach

Behind the Blog: Rare Books and Baseball Brain

404 Media·Aug 7, 15:43 UTC · Aug 7, 2026Industry

Week in review: Salesforce and ServiceNow portals exposed for 17 months, exploited Metabase 0-day

Weekly digest: exploited Metabase zero-day breached Framework; Salesforce/ServiceNow portals read for 17 months; Microsoft patched 400+ flaws.

Help Net Security's week in review aggregates top stories: a 'City-Forum' campaign tracked by Reco has been pulling records from Salesforce and ServiceNow portals worldwide for 17 months, and Framework suffered a breach via an exploited Metabase zero-day exposing customer contact and IP data. It also covers Microsoft's August 2026 Patch Tuesday fixing 400+ flaws including exploited zero-day CVE-2026-68820, Cisco's fix for exploited firewall DoS bug CVE-2026-20349 (added to CISA KEV), and a second N-able N-central hotfix for actively exploited CVE-2026-18577. Other items include GitHub expanding Dependabot malware alerts to eight package ecosystems and EU AI Act enforcement beginning on 2 August 2026.

Help Net Security·Aug 16, 00:00 UTC · Aug 16, 2026Industry in the wildCVE-2026-18577CVE-2026-68820CVE-2026-203491

Does Syntax Matter? A Graph-Augmented Variational Topic Model for Computational Social Sciences

SCPTM graph-augmented variational topic model shows syntax aids topic diversity and descriptor quality but gains stem mainly from the variational encoder.

The Structural Contextual Probabilistic Topic Model represents corpora as heterogeneous document-word graphs with lexical and syntactic edges processed by a Graph Attention Network inside a VAE for mixed-membership topic distributions. Across four corpora, neural gains in document-topic alignment are attributable to the variational encoder rather than syntax, while graph-augmented variants improve topic diversity everywhere. Dependency paths add value on argumentative deliberative texts but are redundant in technical and institutional registers.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 17:39 UTC · 9d agoAI research1

2026 Cyber Insurance Trends Report: What's Changed and What You Need to Know

Huntress survey: CIRCIA reporting mandates now live, BEC claims exceed ransomware, exfiltration-heavy attacks cost twice as much, premiums rising.

Huntress's 2026 cyber insurance trends report, based on its own survey, finds 79% of respondents carry cyber insurance while 58% report shrinking coverage over five years. New CIRCIA federal reporting mandates and EU NIS2 requirements are reshaping policies, business email compromise now drives more claims than ransomware, and data exfiltration has replaced encryption as the dominant ransomware tactic at roughly twice the cost. After three years of declining premiums, rates are climbing again, and most businesses now refuse to pay ransoms.

Huntress·Sep 1, 17:00 UTC · 15d agoIndustry1

Before You Poll with LLMs: A Deliberative Diagnostic Framework

Deliberative diagnostic shows all five tested frontier LLMs misrepresent human belief shifts after arguments, with GPT-5.1 reversing on outgroup questions.

The Deliberative Polling Diagnostic Framework compares human and LLM persona belief shifts after identical informational interventions, using data from America in One Room (526 personas, 72 questions). All five frontier models tested failed uniquely: GPT-5.1 exhibited partisan reversal (80% on outgroup vs 26% on policy questions), Gemini 2.0 Flash, Claude Sonnet 4.5 and Llama 3.3 70B overshot at 5-7x human magnitude, and DeepSeek V3 showed near-zero change (rigidity). The authors term the underlying signature 'self-sycophancy', conformity to the model's internal persona stereotype rather than reasoning from provided information.

arXiv cs.AI / cs.LG / cs.CL·Sep 14, 16:41 UTC · 2d agoAI research

Safety for Whom? Refusing the Right Subset of a Topic, Not the Whole Topic

Multiverse Computing's Hugging Face post argues language models should refuse only the relevant subset of a topic instead of over-refusing whole subjects.

A Hugging Face blog post by Multiverse Computing examines refusal granularity in language models, arguing models should refuse the relevant subset of a topic rather than the entire topic. No full article text was available for additional technical detail.

Hugging Face Blog·Sep 8, 14:23 UTC · 8d agoAI safety & security

Reason Through the Latent! Making Latent Visual Reasoning Necessary

Researchers introduce CVRR, forcing multimodal models to rely on recurrent latent computation rather than accessible image tokens, validated via causal interventions and benchmarks.

The paper presents Causal Visual Recurrent Reasoning (CVRR), which makes recurrent hidden-state computation the required image-conditioned path for prediction in vision-language models. Before decoding, visual states and the original multimodal KV cache are removed so only the final recurrent state carries image information to the answer. CVRR retains strong performance on V*, MMVP, BLINK, and MME-RealWorld-Lite while comparable latent reasoners fail under the same constraint. Causal interventions show predictions remain sensitive to recurrent content and that persistent visual evidence causally revises the recurrent trajectory.

Hugging Face daily papers·Sep 5, 20:00 UTC · 10d agoAI research

Making Knowledge Distillation Cheap Enough to Run at Scale

Hugging Face blog by Multiverse Computing describes techniques making knowledge distillation cheap enough for large-scale training.

A Hugging Face blog post from Multiverse Computing (CAI) presents methods for reducing the cost of knowledge distillation so it can be run at scale. The post is aimed at practitioners compressing large models into smaller, cheaper ones for production use.

Hugging Face Blog·Aug 10, 10:05 UTC · Aug 10, 2026AI research

An open letter to Dario: if you mean it, open the weights

Open letter urges Dario Amodei to champion a law forcing all publicly released AI models to ship as open weights, arguing regulation otherwise gets captured.

A blogger responds to Anthropic's essay 'We Must Pace the Frontier' by proposing a law requiring any AI model offered to the public be released as open weights, citing Anthropic's 'Fable' and OpenAI's 'Astra' as examples. The author argues frontier lab valuations depend on proprietary weights, so mandatory openness would cut funding for future training runs and slow progress across all labs. The letter contends regulatory approaches like embedded evaluators, compute thresholds, and antitrust waivers inevitably lead to regulatory capture favoring incumbents. It appeals to Amodei's record on safety, export controls, and Anthropic's PBC structure to back the proposal.

Hacker News · AI·Sep 12, 19:15 UTC · 3d agoAI policyHN 40↑ · 1 comments1

Real-Time Intelligence with IBM Time Series Models on Confluent

IBM Research post on the Hugging Face blog describes running IBM time series models on Confluent for real-time intelligence.

Hugging Face's blog published an IBM Research post titled 'Real-Time Intelligence with IBM Time Series Models on Confluent.' No article body was available for classification, so details are limited to the title. The title indicates guidance on deploying IBM time series models alongside Confluent streaming infrastructure for real-time analytics.

Hugging Face Blog·Sep 2, 13:49 UTC · 14d agoAI tools & infra

Evaluating Time-Series Foundation Models and Multimodal Dietary Context for CGM Forecasting

Study finds zero-shot time-series foundation models underperform on CGM forecasting; fine-tuned Chronos-Bolt cuts RMSE up to 18.4% and dietary context adds signal.

The paper evaluates time-series foundation models for continuous glucose monitoring forecasting across eight public datasets covering Type 1 diabetes, Type 2 diabetes, and non-diabetes populations. Under a unified protocol, zero-shot foundation models did not consistently outperform baselines like Elastic Net and PatchTST, but lightweight fine-tuning did, with fine-tuned Chronos-Bolt reducing RMSE by 6.5%-18.4% in the T1D cohort and 8.6%-18.2% in the non-diabetes/T2D cohort. A residual-based fusion framework adding dietary context from CGMacros reduced overall RMSE by about 3% and postprandial RMSE by about 15% versus CGM-only baselines.

arXiv cs.AI / cs.LG / cs.CL·Sep 10, 17:43 UTC · 6d agoAI research

RESCUE-BENCH: Towards Relation-Aware Multi-Party Emotional Support Conversation Systems

Researchers introduce RESCUE-Bench, a video benchmark of 191 couple and family conversations evaluating LLMs on relation-aware multi-party emotional support.

RESCUE-Bench is built from real couple and family interview conversations, containing 191 samples, 7,079 annotated turns, and 1,064.8 minutes of video. It defines six tasks measuring two capabilities: Relational Understanding and Relation-Sensitive Support. Experiments with ten LLMs show models handle local emotional cues but struggle with relation pattern prediction, viewpoint prediction, and support strategy prediction.

Hugging Face daily papers·Sep 8, 20:00 UTC · 7d agoAI research

Diffusion Models and Concept Formation

Paper argues diffusion models implicitly form Cobweb-like concept hierarchies, with a basic level emerging at intermediate noise levels.

The authors draw a formal correspondence between diffusion models and Cobweb, a classic incremental concept-hierarchy learner, noting both are hierarchical Bayesian density models with Gaussian prototypes. Modes of the diffusion model's noisy marginals form a hierarchy whose basic level sits at intermediate noise, where class identity commits. The correspondence is tested on MNIST and Fashion-MNIST via mode-finding. Diffusion is reframed as a cognitive model of concept formation.

arXiv cs.AI / cs.LG / cs.CL·Sep 11, 16:43 UTC · 5d agoAI research

Quantifying the Engagement Trap: Impact of Short-form Video Recommender Systems on Users with ADHD

A 302-participant study finds engagement-optimized short-form video recommenders cause disproportionate time blindness and distress for users with ADHD.

Researchers ran a stratified Prolific study with 302 participants comparing short-form video recommendation experiences with and without ADHD. Participants with ADHD reported significantly higher time blindness, post-usage regret, and emotional distress despite perceiving recommendations as similarly relevant. The paper proposes neurodiversity-aware, human-centered design interventions to mitigate these algorithmic harms.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 17:38 UTC · 9d agoAI research

Another Blueprint In The Wall: How to Ask Frontier AI Like a Kid?

Six frontier models from OpenAI, Anthropic, xAI, and Google DeepMind converge on one imagined successor architecture when asked under a school-audience framing.

Researchers ran ten independent sessions per model type across six frontier models using a three-stage prompt sequence progressing to a full ASCII backbone architecture. Under school-audience framing, responses repeatedly converged on a shared motif including persistent latent state, adaptive computation, memory, specialist routing, verification, and stopping control, while control runs without the framing produced heterogeneous responses. A GPT-5.6 Sol output closely overlapped an architecture independently sketched by GPT-6 Astra, raising questions about shared design priors or motif propagation between model families. The paper coins 'epistemic jailbreak' for the observed loss of provenance discipline as prompt specificity increases.

Hugging Face daily papers·updated 13:01 UTC · 6h ago·first Sep 12, 20:00 UTC · 3d agoAI research 11 sources

What the AI Warning Letter Completely Missed

Opinion piece argues the recent AI warning letter identifies a risk window but omits which actors pose risks and who can mitigate.

This Dark Reading commentary critiques a recent AI warning letter for correctly identifying an approaching risk window while failing to name who is coming through it or who will close it. The piece is brief opinion commentary on AI risk discourse rather than a technical report.

Dark Reading·Sep 3, 17:23 UTC · 13d agoAI safety & security

Microsoft says virtually nobody was grabbing NYT articles through its chatbot

Microsoft files summary judgement briefs in NYT copyright lawsuit, arguing only 59,545 of 8.2M Copilot logs show substantial overlap with news content.

Microsoft filed new legal filings in the consolidated copyright lawsuit brought by The New York Times, authors, and publishers against Microsoft and OpenAI, arguing for summary judgement. Analysis of 8.2 million Copilot chat logs found only 59,545 contained at least 16 words in common with news content, with just 24 responses showing at least 30 matching words in the authors' case. Microsoft argues the numbers support fair use, claiming Copilot rarely reproduces substantive chunks; the Times says discovery shows Microsoft and OpenAI stole from it, and the Trump administration filed a statement of interest supporting OpenAI.

The Verge · AI·Sep 4, 16:05 UTC · 12d agoAI policy1

Podcast: Humans Are Reading Your ChatGPT Conversations

404 Media podcast examines contractors who read real ChatGPT user conversations, plus Automattic CEO turmoil and a16z's enshittification essay.

404 Media's weekly podcast leads with an investigation into 'Project Lily,' the human contractors who review real ChatGPT users' prompts and conversations; the reporter reviewed internal documents about the program. It also covers leadership upheaval at Automattic, where CEO Matt Mullenweg was put on leave of absence and then claimed he was back in control. A subscriber-only segment discusses a16z's argument that enshittification is not real.

404 Media·14:19 UTC · 4h agoAI safety & security

TempCloze: Can Video-LLMs Identify the Missing Middle?

TempCloze benchmark tests Video-LLMs' temporal reasoning with 1,521 videos, finding temporal alignment is the primary failure mode across 31 models.

TempCloze is a video cloze benchmark in which models must identify the true missing middle clip given the beginning and ending clips, using 1,521 carefully filtered videos from seven sources, mostly long-take and egocentric footage. Distractors are constructed along three dimensions: Semantic, Alignment and Progression, with shared scenes and objects to reduce appearance cues. Evaluation of 10 proprietary and 21 open-source Video-LLMs found Alignment is the primary bottleneck, with models often recognizing plausible semantics and local event progression but struggling with temporal alignment.

Hugging Face daily papers·Aug 31, 20:00 UTC · 15d agoAI research

Omni-Streaming Thinking

Omni-Streaming Thinking fixes premature cross-modal commitment in streaming omni-modal models via pending claims verified against modality-specific evidence, beating baselines by over 10%.

The paper identifies 'premature cross-modal commitment', where streaming models keep relaying early visual interpretations even after audio contradicts them. OST generates evidence-linked pending claims with future verification intervals, stores audio and visual evidence separately, and refutes claims when contradictory evidence appears. Built on a frozen Qwen3-Omni-30B-A3B-Instruct backbone with lightweight adaptation, it outperforms open baselines by more than 10% relative on five streaming and audio-visual benchmarks. On the new OST-DiagBench it reaches d-prime 2.95 versus at most 1.38 for open baselines, while reducing vision-induced auditory hallucinations.

Hugging Face daily papers·Sep 13, 20:00 UTC · 2d agoAI research1

If the Markets Reject OpenAI and Anthropic, the US Should Nationalize Them

Opinion essay argues the US should nationalize OpenAI and Anthropic into public labs if markets reject their trillion-dollar IPO valuations.

Sanders and Schneier argue in The Guardian that OpenAI and Anthropic may never be sustainably profitable, citing commodity models, short depreciation windows, and free open-source competitors only months behind in capability. They propose converting the labs into US national labs or regulated public utilities if markets reject their recently filed IPOs, which buzz valued at trillions of dollars. They cite public backlash to AI datacenters, Nvidia's slumping stock, and public AI labs in Switzerland, Spain, and Singapore as context.

Schneier on Security·Aug 15, 00:00 UTC · Aug 15, 2026AI industry

From Reweighting to Rewriting: Unlocking the Intervention Effects of Influential Samples in Training Data Attribution

Study shows rewriting responses of influence-selected training examples shifts LLM behavior more strongly than reweighting the same samples.

The paper examines training data attribution, arguing that influence functions identify high-leverage examples whose value goes unrealized under conventional weight-based reweighting interventions. It introduces influence-guided response rewriting, which replaces the responses of influence-selected examples with behavior-aligned or behavior-opposed supervision while keeping instructions fixed, tested across four open-weight LLMs using epistemic abstention as the primary testbed. Rewriting produces stronger, more persistent, and bidirectional behavioral shifts, including on safety refusal, while reweighting the same examples yields weak, inconsistent effects. The results motivate intervention-aware evaluation of TDA methods.

Hugging Face daily papers·Sep 1, 20:00 UTC · 14d agoAI research

Verifiable Social Reasoning for LLM Assistants

Fuse, a multi-agent simulation with hidden motives, evaluates LLM social reasoning, revealing compounding difficulty from user mediation and bias sensitivity.

Fuse is a multi-agent simulation framework in which a target agent with a hidden motive interacts with other agents including one representing the user, who consults the evaluated assistant to infer the motive, providing verifiable ground truth by construction. Simulation faithfulness is validated through a human study with 24k annotations. Applied to 12 LLMs, it shows user mediation compounds social reasoning difficulty, models are systematically sensitive to biased user framing, models may need more details than humans, and longer conversations do not always improve performance. The framework and a 21k-example dataset are open-sourced.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 17:37 UTC · 1d agoAI research

The Frontier AEO Tracker: What Astra Chooses (and every other frontier model, and what you can do about it)

Latent Space launches an AEO tracker scoring 7 frontier models' product recommendations across 161 categories, revealing generational bias flips.

Latent Space built a tracker measuring Answer Engine Optimization by running 6 prompt variations across 7 frontier models with search enabled over 161 product categories, scoring first choices, alternatives, mentions, and anti-recommendations. It found 28 categories with a universally dominant primary choice and observed soft biases, such as models favoring their own lab's coding agents. Analysis of Anthropic's Sol→Astra and Opus→Fable generations showed newer models consulting fewer sources and being less likely to change answers when questions are paraphrased.

Latent Space·Sep 7, 21:32 UTC · 8d agoAI research