All entries

Tensorizing Engram: Sharing Latents Across N-Gram Embeddings is Beneficial in LLMs

arXiv:2606.08347v1 Announce Type: cross Abstract: Modern language models represent text using discrete token-level embeddings, which forces recurring multi-token patterns to be learned implicitly acro

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Industry

Tesla AI chip design engineering reviews are so great! Team is awesome. Our AI6 chip might set a record for most amount of usable intelligen…

DGX agent

Elon Musk posted on X about Tesla's AI chip design engineering team and their work on the AI6 chip, expressing confidence in the team's performance and suggesting the chip could achieve notable capabi

industryelon-musk--x

9 Jun 2026

Test-Time Adaptive Composition for Machine Learning as a Service (MLaaS) in IoT Environments

arXiv:2606.07685v1 Announce Type: cross Abstract: The dynamic nature of Internet of Things (IoT) environments affects the long-term effectiveness of Machine Learning as a Service (MLaaS) compositions.

researcharxiv-cs-ai

Test-Time Scaling in Multimodal Foundation Models: A Comprehensive Survey of Generation and Reasoning

arXiv:2606.08231v1 Announce Type: new Abstract: Test-time Scaling (TTS) has emerged as a pivotal research direction for enhancing model performance by dynamically allocating computational resources du

researcharxiv-cs-cv

Testing the Black Box: Structural Barriers to Independent Evaluation of Consumer-Facing Health LLMs

arXiv:2606.08483v1 Announce Type: new Abstract: Background: Consumer-facing large language models are now a common source of health information, and they interpret and personalize responses rather tha

The ACUTE Protocol: Operationalizing Language Model Activations for Better Calibration, Utility, and Trust

arXiv:2606.07822v1 Announce Type: cross Abstract: As language models improve and become increasingly deployed to solve a variety of tasks, trustworthiness becomes essential. Calibration is a good prox

The AI Epistemic Deference Index: A Continuous Measure of Sycophancy

arXiv:2606.07897v1 Announce Type: new Abstract: Current AI models frequently exhibit epistemic sycophancy, endorsing claims to agree with a user. Existing evaluations typically measure this either by

The AI supersystem shift: Why Arista’s 1.6T announcement is an Ethernet inflection point

The networking industry loves inflection points. Over the years, we have had many new compute models that require the network to evolve. For as long as I can remember, the holy war between InfiniBand

industrysiliconangle

The best AI infrastructure shouldn't be reserved for the biggest companies. Together AI is partnering with @pax8 to bring powerful, cost-eff…

applicationstogether-ai--x

The best AI infrastructure shouldn't be reserved for the biggest companies. Together AI is partnering with @pax8 to bring powerful, cost-efficient AI and leading open-source models to small and mid-si

Tutorials

The CIFAR Synthetic Evidence Corpus for Detecting AI-Generated Evidence

arXiv:2606.07916v1 Announce Type: new Abstract: The growing ability of generative models to produce realistic documents poses a direct challenge to evidentiary workflows in the justice system and the

tutorialsarxiv-cs-ai

The Confidence Trap: Calibration Attacks for Graph Neural Networks

arXiv:2606.08467v1 Announce Type: cross Abstract: While confidence calibration is essential for trustworthy decision-making in safety-critical applications, the robustness of calibrated GNNs to advers

Agents

// The Consistency Illusion // Multi-agent debate can make agents agree on the final answer while their underlying reasoning stays misaligne…

// The Consistency Illusion // Multi-agent debate can make agents agree on the final answer while their underlying reasoning stays misaligned. This work finds that consensus on the output hides disagr

agentsdair-ai--x

The Cross-Architecture Substrate: A Domain-Transcendent, Calibration-Surviving Geometric Invariant of Modern Vision Encoders

arXiv:2606.07882v1 Announce Type: cross Abstract: Different vision neural networks -- trained to classify, contrast, reconstruct, or match images to text -- should have correspondingly different inter

The Download: whole-body rejuvenation drugs and five things to know about AI

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. David Sinclair plans to test whole-body rejuvenation drugs in

researchmit-tech-review

The Easy, the Hard, and the Learnable: Confidence and Difficulty-Adaptive Policy Optimization for LLM Reasoning

arXiv:2606.07950v1 Announce Type: new Abstract: RL with verifiable rewards can substantially improve LLM reasoning, yet standard GRPO-style training often treats easy, hard, and learnable questions al

safetyarxiv-cs-lg

The EU warns that AI-boosted chemical synthesis is helping European drug gangs develop new 'designer' drug precursors that evade existing product blacklists (Michael Peel/Financial Times)

Michael Peel / Financial Times: The EU warns that AI-boosted chemical synthesis is helping European drug gangs develop new “designer” drug precursors that evade existing product blacklists — EU agency

industrytechmeme

The evidence of two tier policing and judiciary is clear for all to see.

This post claims that there is evident disparity in how the police and judicial system treat different groups or individuals. The assertion suggests systemic inconsistency in law enforcement and court

industryelon-musk--x

The fact that Anthropic may take away subscription access to Fable in two weeks is weird & discourages investing in learning about the model…

applicationsethan-mollick--x

The fact that Anthropic may take away subscription access to Fable in two weeks is weird & discourages investing in learning about the model. Subscription use is how you figure out what the model is g

The Flexibility Trap: Rethinking the Value of Arbitrary Order in Diffusion Language Models

arXiv:2601.15165v4 Announce Type: replace-cross Abstract: Diffusion Large Language Models (dLLMs) break the rigid left-to-right constraint of traditional LLMs, enabling token generation in arbitrary o

The Governance of Human-LLM Interaction: Safety Gating, Civility Steering, and Affective Default Lock-In

arXiv:2606.08172v1 Announce Type: cross Abstract: Large language models (LLMs) increasingly mediate high-stakes interactions in finance, medicine, and mental-health support, yet users have limited con

The Hidden Bias of Process Reward Models:PRISM for Rewarding the Right Reasoning

arXiv:2606.09078v1 Announce Type: new Abstract: Process Reward Models (PRMs) improve credit assignment for reasoning by providing step-level feedback. However, we identify a hidden bias in PRMs caused

safetyarxiv-cs-lg

The Injection Paradox: Brand-Level Suppression in Safety-Trained LLM Recommendations via RAG Context Injection

model-releasesarxiv-cs-lg

arXiv:2606.09204v1 Announce Type: new Abstract: We present a reproducible failure mode of safety training in RAG-based LLM recommendation -- the Injection Paradox -- in which prompt injections embedde

The Label Horizon Paradox: Rethinking Supervision Targets in Financial Forecasting

arXiv:2602.03395v4 Announce Type: replace Abstract: While deep learning has revolutionized financial forecasting through sophisticated architectures, the design of the supervision signal itself is rar

The Last Visible Pixel: Probing Fine-Scale Perception in Vision-Language Models

arXiv:2606.07861v1 Announce Type: cross Abstract: Recent vision-language models (VLMs) excel at multimodal understanding and reasoning, yet their fine-grained visual perception remains underexplored.

The Mirrored Influence Hypothesis: Efficient Data Influence Estimation by Harnessing Forward Passes

arXiv:2402.08922v3 Announce Type: replace Abstract: Large-scale black-box models have become ubiquitous across numerous applications. Understanding the influence of individual training data sources on

The Montparnasse Algorithm for RNA Design

arXiv:2606.07562v1 Announce Type: cross Abstract: RNA design consists of discovering a nucleotide sequence that optimizes predefined criteria, such as secondary structure. It is useful for synthetic b

Local Ai

The Need for Neural ISP in the Small-Pixel Era: How Shrinking Pixels Push Optics to the Limit and Neural Restoration Pushes Back

arXiv:2606.07675v1 Announce Type: cross Abstract: Smartphone telephoto cameras are approaching a 'telephoto physics wall': as pixel pitches shrink toward sub-0.5 micron, the optics remain limited by g

local-aiarxiv-cs-cv

The New York Times published a roundtable discussion between @DAcemogluMIT, @deanwball, @clarashih & myself about the future of AI & who win…

applicationsethan-mollick--x

The New York Times published a roundtable discussion between @DAcemogluMIT, @deanwball, @clarashih & myself about the future of AI & who wins at work. I think it is a really nice overview of the core

Tutorials

The Routing Plateau: Understanding and Breaking the Accuracy Limits of LLM Routers

arXiv:2606.07587v1 Announce Type: new Abstract: LLM routing has become a popular approach to improve the cost-quality trade-off of LLM services by dynamically selecting a model for each query. Recent

tutorialsarxiv-cs-lg

The Sample Complexity of Parameter-Free Stochastic Convex Optimization

model-releasesarxiv-cs-lg

arXiv:2506.11336v2 Announce Type: replace Abstract: We study the sample complexity of stochastic convex optimization when problem parameters such as the distance to optimality and the Lipschitz consta

The Spectral Dynamics and Noise Geometry of Muon

arXiv:2606.08388v1 Announce Type: new Abstract: Muon replaces a matrix gradient G=USigma V^op by its polar factor UV^op. This keeps the singular directions selected by the gradient, but makes the upda

safetyarxiv-cs-lg

Agents

The Token Not Taken: Sampling, State, and the Variability of AI Agent Outputs

arXiv:2606.08998v1 Announce Type: new Abstract: Agentic AI systems can behave differently across runs: the same request may produce a different plan, a different tool call, a different code edit, or a

agentsarxiv-cs-ai

The Topological Dual of a Dataset: A Logic-to-Topology Encoding for AlphaGeometry-Style Data

arXiv:2604.18050v2 Announce Type: replace Abstract: AlphaGeometry represents a milestone in neuro-symbolic reasoning, yet its architecture faces a log-linear scaling bottleneck within its symbolic ded

researcharxiv-cs-ai

The truth is that there are VASTLY more hate crimes, especially aggravated rape and murder, per person by Blacks against Whites than the oth…

The truth is that there are VASTLY more hate crimes, especially aggravated rape and murder, per person by Blacks against Whites than the other way around. The is not remotely debatable, as the numbers

industryelon-musk--x

The US FCC waives its deadline for Amazon to deploy half of its Leo satellites by July; Amazon is still required to launch all 3,232 satellites by July 30, 2029 (Michael Kan/PCMag)

Michael Kan / PCMag: The US FCC waives its deadline for Amazon to deploy half of its Leo satellites by July; Amazon is still required to launch all 3,232 satellites by July 30, 2029 — The Federal Comm

applicationstechmeme

The Value of Personalized Recommendations: Evidence from Netflix

arXiv:2511.07280v5 Announce Type: replace-cross Abstract: Personalized recommendation systems shape much of user choice online, yet their targeted nature makes separating out the value of recommendati

TheoremBench: Evaluating LLMs on Theorem Proving in Formal Mathematics

arXiv:2606.09450v1 Announce Type: new Abstract: LLMs have recently achieved strong results on formal proving benchmarks. However, existing evaluations remain heavily concentrated on competition-style

Theoretical Foundations of Continual Learning via Drift-Plus-Penalty

model-releasesarxiv-cs-lg

arXiv:2606.08452v1 Announce Type: new Abstract: In many real-world settings, data streams are nonstationary and arrive sequentially, requiring learning systems to adapt continuously without retraining

There has been a lot of hand wringing on the appropriate valuation of SpaceX. Some large institutions believe SpaceX can only be valued at h…

model-releaseselon-musk--x

There has been a lot of hand wringing on the appropriate valuation of SpaceX. Some large institutions believe SpaceX can only be valued at half what the market seems to be willing to pay for it. Other

They call for calm in the face of horror. This is inhuman and manipulative. The proportionate and natural response to an horrific incident i…

They call for calm in the face of horror. This is inhuman and manipulative. The proportionate and natural response to an horrific incident is disgust, fear and anger. They want to downplay the horror

industryelon-musk--x

Tutorials

They didn’t mean pause AI research, they meant pause your AI research

tutorialsjeremy-howard--x

Jeremy Howard argues that calls to pause AI research are selectively applied, with restrictions primarily targeting independent researchers while well-resourced labs continue development, creating an

They ruled Iryna’s killer is incompetent to stand trial. The same system ruled this man was plenty competent enough to be released back into…

model-releaseselon-musk--x

They ruled Iryna’s killer is incompetent to stand trial. The same system ruled this man was plenty competent enough to be released back into society dozens of times. It’s past time to remove these lef

Think Before You Act: Intention-Guided Reasoning for LLM-Based Location Prediction

arXiv:2606.08122v1 Announce Type: new Abstract: Predicting a user's next Point-of-Interest (POI) based on their historical check-in records is a fundamental task in location-based services. While rece

Thinking-Based Non-Thinking: Solving the Reward Hacking Problem in Training Hybrid Reasoning Models via Reinforcement Learning

arXiv:2601.04805v2 Announce Type: replace Abstract: Large reasoning models (LRMs) have attracted much attention due to their exceptional performance. However, their performance mainly stems from think

Local Ai

Thinking Without Images: Internalizing Visual Manipulation with On-Policy Self-Distillation

arXiv:2606.08719v1 Announce Type: new Abstract: ''Thinking with Images'' has emerged as an effective paradigm for fine-grained visual reasoning: by explicitly zooming into relevant regions and reasoni

local-aiarxiv-cs-cv

This is a super exciting release - Claude Fable 5 is the same underlying model as Mythos but with added safeguards. The benchmarks are great…

model-releaseskarpathy--x

This is a super exciting release - Claude Fable 5 is the same underlying model as Mythos but with added safeguards. The benchmarks are great and it's SOTA on everything by a margin but I'll add that *

This is actually a pretty measured bet compared to spending of the U.S companies. OpenAI alone is about twice as much. If things falls apart…

This is actually a pretty measured bet compared to spending of the U.S companies. OpenAI alone is about twice as much. If things falls apart, the US will be hit harder. 🚨BREAKING: CHINA IS AGI-PILLED

safetygary-marcus--x

Agents

this is sick haha.

this is sick haha. The Agent Open 🎾🏓 Everyone loves pickleball. We’re hosting a massive pickleball tournament during the AI Engineer World Fair. I’m so excited to see this event come together. This is

agentsjerry-liu--x