The Archive

As video generation models like Veo 3.1 and LTX-2 advance, their ability to accurately represent diverse global cultures remains a critical yet understudied frontier. Current metrics, such as VideoScore, only measure visual quality but offer no mechanism for assessing cultural faithfulness. Consequently, a model that replaces a Namaste with a handshake receives the same score as one that generates the gesture correctly. We propose CultureScore, a compositional evaluation framework that decomposes cultural faithfulness into three granular dimensions: Identity (who is represented), Context (cul...

Instruction-following audio language models (ALMs) can be augmented with explicit acoustic cues, yet it remains unclear whether such cues are used in a grounded way when the raw audio is already available. We study this question in speech emotion recognition (SER) by deriving six interpretable acoustic concept tokens from the standardised eGeMAPS paralinguistic feature set. These tokens summarise energy, pitch, dynamics, brightness, formants, and voice quality, and are appended to the textual prompt while the audio input is kept unchanged. Across the widely used FAU-Aibo and IEMOCAP benchmark...

While the AI fundraising machine keeps breaking its own records, some founders are building in the other direction. Mirror founder Brynn Putnam just raised money for Board, a startup focused on bringing people together through in-person games and social experiences. Cyberdeck creators are going viral crafting whimsical DIY computers that literally encourage users to touch grass. Unlike the AI-free browser crowd, this doesn’t just feel like backlash, […]

When Quilty hit the industry trades earlier this year, the AI startup promised that its tool could accurately predict a film's success just by reading the script. When people actually got a chance to experiment with Quilty's product, though, they were left skeptical. Even with all the available data in the world, it predicted the script for Christy, which would go on to be a box office flop, would outperform the script for Sinners, which became an Oscar-winning blockbuster. As many AI execs have pitched before, Quilty's founders believe that can help "democratize" their industry by giving up-...

The Australian data center operator plans to set up 5GW of capacity in India.

We will no longer accept public pull requests. [...] A substantial patch used to imply substantial effort, and that effort was a reasonable proxy for good faith. That assumption no longer holds. [...] Whether code was typed by hand is beside the point. What matters is who is responsible for it once it enters the browser. Ladybird is becoming a browser for real users. The people introducing changes to it must be the people who decide those changes belong in the project, and who will answer for the consequences. — Andreas Kling , Changing How We Develop Ladybird Tags: ladybird , ai-ethics...

On June 5, 404 Media reported that attackers had been using Meta’s AI customer support agent to steal Instagram accounts. Their approach was simple: They asked the agent to link the accounts to email addresses that they controlled, and the agent complied. One attacker broke into the dormant Obama White House account and made pro-Iran…

a quiet day

In the current environment, remaining heads down has diminishing returns; at some point, you have to make some noise just to remind the market you exist.

Navigate AI safety with confidence. This guide explains 7 key enterprise AI safety themes to help you address biases, misinformation, and other real-world risks.

AI enthusiasts are in a race against time, AI skeptics are in a race against entropy Charity Majors neatly captures the dynamic between AI enthusiasts and AI skeptics, both of whom are trying to build great software, often in the same teams: The enthusiasts are not wrong . We are starting to see real, non-imaginary, discontinuous leaps in capabilities from teams that lean in hard to working with AI. And this does not feel like a normal technology cycle where you can wait for the dust to settle; teams that sit this out while competitors are hustling could be out of business before the dust set...

The AI giant's co-founder explained why the company may tap the public market for capital and why tokenmaxxing pushback isn't a concern.

The Airbnb CEO said last year it hasn't struck an LLM partnership because existing products weren't quite ready.

Robot demonstrations can distort public perceptions of robotic capabilities.

With just two weeks to go, StrictlyVC Los Angeles is quickly approaching. On Thursday, June 18, at The Aerospace Corporation Campus in El Segundo. Investors, founders, and tech leaders will gather for an evening of conversations exploring some of the most consequential shifts taking place across venture capital, defense technology, artificial intelligence, and advanced industry. Secure your spot here. For […]

Estonian government benchmark shows how dozens of models combat Russia's "strategic narratives."

We talk with the VendingBench authors on evaling Claudes from Haiku to Mythos, and how they build leading, and lasting, frontier evals from scratch.

Musk can't be trusted to protect X user privacy, public commenters warn FTC.

Meta may have one found one way to slash its massive data center bill: tents.

Poke, the startup that lets people use AI agents through simple text messages, has become the first AI agent approved for Apple’s Messages for Business platform.

Kevin O'Leary agreed to halve the size of his planned 40,000-acre data center in Utah amid mounting pressure from residents and activists, as reported earlier by local affiliate ABC4. The Shark Tank star sent a letter to Utah Senate President J. Stuart Adams on Thursday, saying that he will remove 19,430 acres from the project, located in and around the Locomotive Springs Waterfowl Management Area. The change comes just days after Adams called on O'Leary to slash the size of his Project Stratos data center by 75 percent, which would reduce it to about 10,000 acres. Adams also asked O'Leary to...

Parameter-efficient finetuning methods based on spectral decomposition have enabled progress in Continual Learning. In this paper we introduce TailLoR, which utilizes the singular bases U and V of the pre-trained weights as a fixed reference frame to learn a low-rank update applied to the singular value matrix. A soft spectral penalty discourages updates aligned with dominant singular directions, reducing interference while routing fine-grained adaptation into the highly flexible, long-tail spectral coordinates.

For a humanoid robot to be deployed in the real world, the choice of command space (i.e., the interface between task planning and whole-body control) is crucial. Existing whole-body controllers typically demand dense kinematic or spatial references that planners struggle to synthesize from task semantics. We instead propose a compact, explicit interface that is intuitive, general, modular, and expressive enough for diverse manipulation skills. To this end, we introduce HANDOFF, a single humanoid whole-body controller that follows this interface and is distilled via multi-teacher KL distillati...

Code language models need repository-level context to resolve imports, APIs, and project conventions. Existing methods inject this knowledge as long inputs (retrieved through RAG or dependency analysis) or through per-repository fine-tuning and LoRA -- costly at repository scale and brittle to evolving codebases. We introduce Code2LoRA, a hypernetwork framework that generates repository-specific LoRA adapters, effectively injecting repository knowledge with zero inference-time token overhead. Code2LoRA supports two usage scenarios: Code2LoRA-Static converts a single repository snapshot into a...

Robot manipulation alternates between low-risk transit phases that call for fast execution and high-risk contact stages that demand slow, precise motion. Yet existing Vision-Language-Action models (VLAs) only inherit a single fixed speed from training demonstrations. Prior efforts to accelerate VLAs through model compression, KV-cache reuse, or reinforcement learning only shift the policy from one fixed speed to another, and leave deceleration almost unexplored. We observe that the magnitude of each predicted action already governs how fast the robot moves, opening a direct route to controlla...

In this paper, we study regret minimization in repeated games with \emph{adaptive} opponents who can respond based on histories of play. The standard metric of \emph{external regret} in online learning is known to fail to capture such adaptivity. To account for players' counterfactual reasoning, we introduce {\tt Repeated Policy Regret (RP-Regret)}, a game-theoretic metric that measures the difference between the \emph{realized} and the \emph{best-in-hindsight} accumulated utility when all players can \emph{respond} to the history of play. Compared to existing regret notions in this setting, ...

As AI writing assistants become increasingly integrated into real-world drafting and revision workflows, many documents are no longer purely human-written or AI-generated, but instead result from progressive human-AI co-editing. However, existing AI-text detection benchmarks largely focus on final outputs and provide limited understanding of how AI authorship signals emerge, accumulate, or disappear throughout the revision process. We introduce OpAI-Bench, an operation-guided benchmark for studying progressive human-to-AI text transformation across document, sentence, token, and span granular...

Many real-world competitive systems require multiple decision-makers to act simultaneously under shared constraints, limited information, and repeated interaction, as in auctions, resource allocation, and security competition. We study multi-turn simultaneous bidding as a controlled testbed for such problems and propose DNQ, a solver-in-the-loop equilibrium supervision framework for training bidding agents. DNQ alternates between trajectory collection, critic-based payoff estimation, equilibrium computation, and policy imitation. At each visited state, a shared critic predicts either pairwise...

Training recurrent neural networks (RNNs) requires assigning credit across long sequences of computations. Standard backpropagation through time (BPTT) addresses this problem poorly: it is sequential in time, limiting parallelism, and suffers from vanishing or exploding gradients, making long-range associations difficult to learn. We propose Supervised Memory Training (SMT), a method for training nonlinear RNNs that sidesteps recurrent credit propagation entirely by reducing RNN training to supervised learning on one-step memory transition labels $(m_t, x_{t+1}) \rightarrow m_{t+1}$. SMT acqu...