MLINFO COLLECTION
日次更新の新着・公開記事を新しい順に掲載しています。
This work considers the design of stepsize schedules for stochastic gradient descent in smooth convex optimization. Throughout, we assume that gradient noise is conditionally zero-mean with bounded…
We study whether a language model can help design controllers whose behavior can be formally checked. In proof-carrying evolution (PCE), a search proposes controllers together with their certificates.
Acoustic indices are widely used as rapid proxies for biodiversity, but they summarise the distribution of sound energy rather than the species producing it and often track richness poorly. We revise…
We study the stochastic second-order oracle complexity of solving smooth convex finite-sum optimization with $n$ components. Very recently, Tsedenov \& Kovalev (2026) showed that the minimax optimal…
Hundreds of millions of people now ask large language model (LLM) chatbots for health advice, which raises a question that is simple to pose and hard to answer: does everyone get the same advice? We…
We study how to learn a generative model of clean images using only noisy observations. Diffusion processes and their generalizations to non-Gaussian noise require posterior predictions at every…
Autoregressive video generation enables efficient inference through streaming prediction. However, existing methods are typically restricted to fixed sampling steps, limiting the flexibility to…
Looping re-applies a block of layers so that each token passes through the block more than once. Adding a loop to a trained model is attractive because it leaves the parameter count, and hence the…
Quantum kernel methods are the most concrete near-term proposal for quantum machine learning on classical data, yet published benchmarks disagree on whether they outperform tuned classical kernels,…
Reinforcement learning is a powerful paradigm for training autonomous agents and has achieved impressive performance in complex environments. However, this success often comes at the cost of…
Spline representations provide a compact way to approximate continuous trajectories using a finite number of parameters. When approximating trajectories generated by dynamical systems, the efficiency…
We present GhostPrefix, the first synchronization-free universal adversarial prefix attack against automatic speech recognition (ASR) systems. Unlike prior audio attacks, which are typically crafted…
Large language models (LLMs) can help discover mathematical equations from data, but most LLM-guided symbolic regression methods focus on a single best equation without quantifying uncertainty over…
Large-scale egocentric human videos provide rich and scalable manipulation experience for robot policy learning, yet substantial embodiment gaps in visual appearance, kinematic structure, and action…
The order in which tokens are predicted in language modeling has a profound impact on the quality of generated sequences. Autoregressive and masked diffusion models iteratively fill in tokens and can…
Neural emulators of chaotic PDEs are trained on one-step prediction but deployed autoregressively, where their errors compound. Jacobian-based regularisers aim to curb this growth, but in a chaotic…
This paper presents TurnTurk-GPT, a text-based turn-taking predictor for Turkish obtained by fine-tuning a Turkish GPT-2 model with the TurnGPT objective on synthetic and real dialogue, and evaluates…
Multi-turn negotiation remains a challenging problem for large language model (LLM) agents. While LLMs can conduct natural-language dialogue, existing benchmarks predominantly assess only agreement…
Video generation models synthesize realistic motion, but can they respond correctly to changes in initial geometry? We introduce GeoFlip, a video generation benchmark centered on physical outcome…
As generative artificial intelligence (AI) becomes more capable and widely used, understanding when and how humans should work with these systems is becoming increasingly important. However, the…
When a language model is fine-tuned to follow instructions, or further pretrained on a new subject, it often gets worse at predicting standard web text. This is known as "forgetting." We ask whether…
Recognized as a top threat to AI agents, prompt injection allows instructions from untrusted data (email, documents, websites) to manipulate the agent's operation. The existing best-defended LLM,…
We study high-dimensional graph combinatorial hypothesis testing problems in which a signal is supported on the edge set of an unknown subgraph drawn from a class invariant under vertex relabeling.…
Linear attention scales linearly with sequence length, but its finite-state memory can limit long-context retrieval. In gated linear recurrences, memory retention depends on state decay, which is…