MLINFO COLLECTION
検索拡張生成、文書検索、社内ナレッジQAに関係する論文・実装をまとめます。
Spatial transcriptomics (ST) enables joint characterization of tissue morphology and gene expression. However, 3D volumetric ST acquisition remains expensive.
Fair classification faces a twofold challenge: maintaining high classification accuracy while limiting reliance on sensitive information. Our method FairSplit achieves this by learning a…
On-policy distillation (OPD) is increasingly adopted in the post-training of large language models. However, a stronger teacher may fail to improve the student.
Robust depth estimation aims to maintain high-quality depths across diverse conditions. However, most existing methods estimate depth without taking into account the object-level information.
Data augmentation improves visual recognition by exposing models to synthetic training examples that encourage generalization beyond the observed data. Mixed-sample methods extend this idea by…
Event cameras offer low latency, high dynamic range, and milliwatt-level power consumption, making them ideal for agile robots such as tiny drones navigating through cluttered environments. However,…
Embedding-level scaling expands the capacity of language models through lookup-based memory rather than by increasing model depth, width, or dense computation. Extending this idea to vision-language…
Prediction models degrade when the deployment population differs from the one used to train them, and training data cannot say which feature-response relationships will transfer. Expert elicitation…
Computationally expensive simulations and real-world observations often provide data at a lower temporal resolution than downstream tasks require. Averaging observations at fixed grid points can…
35.2% of task time in world employment is at least partly delegable to AI agents in 2026, and only 2.5% can be handed over in full. Existing measures score exposure to generative AI, mostly for the…
A long-lived assistant that brings up memories without being asked must decide at every turn whether anything in its store is worth an interruption. We show that this decision is a multiple-testing…
Time Series Foundation Models have demonstrated that pretraining on diverse datasets enables effective transfer to new forecasting tasks. Classical methods like ARIMA, however, remain configured and…
Retrieval-augmented generation commonly serves separate models for reranking and answering, duplicating model residency and passage processing. We introduce Glance, which equips the generator with a…
Controllable 3D part generation is pivotal for digital asset creation, yet existing methods lack the flexibility to accommodate diverse user workflows requiring varying levels of control. In this…
Long-text image-text alignment is essential for fine-grained scene understanding, yet existing methods mainly focus on extending context length or improving local alignment. We identify an overlooked…
Automated radiology report generation requires models to do more than produce fluent clinical narratives: they must explicitly state supported diagnoses while avoiding unsupported ones. However,…
Mix-policy optimization incorporates off-policy expert data into the reinforcement learning (RL) loop to transcend the capability boundary of on-policy RL. However, existing methods either provide…
Semi-dense local feature matching commonly aggregates contextual information before coarse-to-fine correspondence estimation. However, intra-image and inter-image aggregation serve different roles:…
Domain generalization (DG) aims to improve the generalization of the model to any unseen domains. Targeting this goal, existing DG methods mainly focus on learning domain-invariant features or…
What does it mean to have an optimal data selection algorithm? Optimality is conventionally framed as finding a subset that minimizes the population error of the classifier trained only on that…
Hallucinated citations make it harder to trace sources and verify the evidence behind scientific claims. Detecting them is difficult because bibliographic records can be incomplete or differ across…
Vietnamese pre-trained encoders have improved rapidly, but the strongest ones are large, multilingual-vocabulary models or depend on external word segmentation, which makes them costly to deploy. We…
Quantitative analysis of environmental phenomena increasingly relies on image-based measurements. In volcanology, the geometric characterization of volcanic plumes and clouds from RGB images remains…
Literature-searching language agents are becoming a standard component of scientific assistants for the life sciences, yet there is little controlled evidence on which parts of their harness matter.…