The University of Osaka

AI agents can now propose atom-level biological hypotheses faster than they can be tested, but trust in an agent-run molecular dynamics simulation rests on a process record, and existing agents leave that record to the language model itself. MDClaw pairs validating tools with a directed acyclic graph (DAG, a graph recording which calculation stages depend on which) that checks stages before execution and records each one; on MDDataBench, which grades against deposited simulations, four open-weight models with MDClaw passed all three runs on 81–100% of 31 test systems. An ablation without the DAG found 11 runs that passed every output check yet had skipped minimization or equilibration, or simulated a different structure.

AI agents can now propose atom-level biological hypotheses faster than they can be tested, but trust in an agent-run molecular dynamics simulation rests on a process record, and existing agents leave that record to the language model itself. MDClaw pairs validating tools with a directed acyclic graph (DAG, a graph recording which calculation stages depend on which) that checks stages before execution and records each one; on MDDataBench, which grades against deposited simulations, four open-weight models with MDClaw passed all three runs on 81–100% of 31 test systems. An ablation without the DAG found 11 runs that passed every output check yet had skipped minimization or equilibration, or simulated a different structure.

A binDCT block transform runs directly on a programmable switch, filtering noisy sensor streams on the forwarding path before in-network machine learning, which raises switch-side classification MCC from 0.35 to 0.42 and matches the server-side value.

A binDCT block transform runs directly on a programmable switch, filtering noisy sensor streams on the forwarding path before in-network machine learning, which raises switch-side classification MCC from 0.35 to 0.42 and matches the server-side value.

Addressing the challenge of jointly modeling instantaneous within-interval causal effects and lagged cross-interval causal effects while accounting for nonstationarity in time-series causal representation learning, this work establishes sufficient conditions for identifying latent states and their instantaneous and lagged causal structures using an observed auxiliary variable such as time or a condition label, and proposes iCReN, a contrastive-learning framework with discrete or continuous auxiliary variables that learns latent representations and estimates both causal structures, with experiments showing accurate recovery of latent states and both structures on synthetic data and utility of the learned representations for downstream forecasting on real-world data.

Addressing the challenge of jointly modeling instantaneous within-interval causal effects and lagged cross-interval causal effects while accounting for nonstationarity in time-series causal representation learning, this work establishes sufficient conditions for identifying latent states and their instantaneous and lagged causal structures using an observed auxiliary variable such as time or a condition label, and proposes iCReN, a contrastive-learning framework with discrete or continuous auxiliary variables that learns latent representations and estimates both causal structures, with experiments showing accurate recovery of latent states and both structures on synthetic data and utility of the learned representations for downstream forecasting on real-world data.

Surveying 172 students at a Japanese national university and analyzing responses with descriptive statistics, paired-samples t-tests, and content analysis, the study compared learners' perceptions of the usefulness, limitations, and future roles of self-access language learning (SALL) services and AI chatbots for English learning, finding that SALL services were rated significantly higher for speaking development, motivation, personalized advice and feedback, writing, and overall contribution, while chatbots were valued for convenience and accessibility, and that learners saw the two as complementary rather than competing resources.

Surveying 172 students at a Japanese national university and analyzing responses with descriptive statistics, paired-samples t-tests, and content analysis, the study compared learners' perceptions of the usefulness, limitations, and future roles of self-access language learning (SALL) services and AI chatbots for English learning, finding that SALL services were rated significantly higher for speaking development, motivation, personalized advice and feedback, writing, and overall contribution, while chatbots were valued for convenience and accessibility, and that learners saw the two as complementary rather than competing resources.

Ogata and Nakashima ask where encoding happens when multimodal language models skip dedicated perceptual encoders and expose a shared transformer to lightly projected patches, audio frames, or discrete visual tokens, and through linear probing, similarities to perceptual encoders, and causal analyses they find the transformer internalizes the missing computation, constructing task-usable perceptual representations within its own early-to-middle layers, a structure they call a Virtual Encoder.

Ogata and Nakashima ask where encoding happens when multimodal language models skip dedicated perceptual encoders and expose a shared transformer to lightly projected patches, audio frames, or discrete visual tokens, and through linear probing, similarities to perceptual encoders, and causal analyses they find the transformer internalizes the missing computation, constructing task-usable perceptual representations within its own early-to-middle layers, a structure they call a Virtual Encoder.