2026-06-08 AI 리서치 브리핑

Code2LoRA: Hypernetwork-Generated Adapters for Code Language Models under Software Evolution에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

VideoKR: Towards Knowledge- and Reasoning-Intensive Video Understanding에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

Video language models (Video-LLMs) are prone to hallucinations, generating plausible but ungrounded content when visual evidence is weak, ambiguous, or biased. Existing methods, such as contrastive decoding (CD), rely on random perturbations to construct contrastive data for hallucination mitigation, but often fail to target the visual cues that drive hallucination or align with model weaknesses. We propose Model-Aware Counterfactual Data based Contrastive Decoding (MACD), an inference strategy that combines model-guided counterfactual construction with contrastive decoding. MACD uses the Video-LLM's own feedback to identify object regions most responsible for hallucination, generating targeted object-level counterfactual inputs rather than arbitrary frame or temporal modifications. These counterfactual inputs are integrated into CD to enforce evidence-grounded token selection during decoding. Experiments on EventHallusion, MVBench, Perception-test, and Video-MME show that MACD consistently reduces hallucination while maintaining or improving task accuracy across diverse Video-LLMs, including Qwen and InternVL, with especially strong gains in scenarios involving small, occluded, or co-occurring objects.

Modern machine learning pipelines leverage large amounts of public data, making it infeasible to guarantee data quality and leaving models open to poisoning and backdoor attacks. Provably bounding model behavior under such attacks remains an open problem. In this work, we address this challenge by developing the first framework providing provable guarantees on the behavior of models trained with potentially manipulated data without modifying the model or learning algorithm. In particular, our framework certifies robustness against untargeted and targeted poisoning, as well as backdoor attacks, for bounded and unbounded manipulations of the training inputs and labels. Our method leverages convex relaxations to over-approximate the set of all possible parameter updates for a given poisoning threat model, allowing us to bound the set of all reachable parameters for any gradient-based learning algorithm. Given this set of parameters, we provide bounds on worst-case behavior, including model performance and backdoor success rate. We demonstrate our approach on multiple real-world datasets from applications including energy consumption, medical imaging, and autonomous driving.

ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time?에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

TIDE: Proactive Multi-Problem Discovery via Template-Guided Iteration에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

We propose Differences in Detection (DnD), an intuitive method to compare two object detection models. Based on the same matching algorithm, it complements the standard metrics of mean Average Precision ($mAP$) and TIDE error analysis with the ability to compare two models directly. More specifically, we calculate the intersection of ground truth labels that are recognized by both models, followed by the corresponding difference sets and the complement set of ground truth labels that are missed by both models. The resulting comparison is more direct and intuitive than a comparison of independent summary statistics. It reveals individual and shared mistakes and becomes particularly interesting when combined with error types. In this case, the differences in detection errors can be analyzed naturally in a standard confusion matrix. While valuable in itself, we believe that one of the best applications of DnD is to guide explainability methods such as ODAM towards metric-relevant examples, grounded in structured subsets. The code for our method is available here: https://github.com/JohannesTheo/differences-in-detection

Scientific observations generate large quantities of unlabeled data which is laborious to hand-label, making unsupervised learning techniques valuable for processing datasets. Among these approaches, contrastive learning provides a convenient mechanism for extracting structural representations from unannotated datasets. For natural imagery, the general approach is to use a variety of data-space augmentation methods in order to generate synthetic samples; however, for scientific observations data-space perturbations can fundamentally alter the underlying data. Our proposed method is to generate contrastive samples by perturbing the network weights rather than the underlying data, thus more closely preserving the structure of the data. We demonstrate this technique using a SimCLR-based pipeline applied over radar observations of meteors, and show performance gains under matched protocols.

Diffusion models excel at generating high-quality, diverse samples, yet they risk memorizing training data when overfit to the training objective. We analyze the distinctions between memorization and generalization in diffusion models through the lens of representation learning. By investigating a two-layer ReLU denoising autoencoder (DAE), we prove that (i) memorization corresponds to the model storing raw training samples in the learned weights for encoding and decoding, yielding localized spiky representations, whereas (ii) generalization arises when the model captures local data statistics, producing balanced representations. Furthermore, we validate these theoretical findings on real-world unconditional and text-to-image diffusion models, demonstrating that the same representation structures emerge in deep generative models with significant practical implications. Building on these insights, we propose a representation-based method for detecting memorization and a training-free editing technique that allows precise control via representation steering. Together, our results highlight that learning good representations is central to novel and meaningful generative modeling.

June 5, 2026 Unlocking dependable responses with Gemini Enterprise Agent Platform’s Agentic RAG Data Management · Machine Intelligence · Natural Language Processing · Product에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

June 4, 2026 Towards passive heart health monitoring via smartphone camera Health & Bioscience · Human-Computer Interaction and Visualization · Machine Intelligence에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

June 3, 2026 The next chapter in flood resilience: Open sourcing Google’s hydrology framework Climate & Sustainability · Earth AI · Generative AI · Open Source Models & Datasets에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

May 28, 2026 A New Era of Discovery: Google Research at I/O 2026 General Science · Generative AI · Global · Quantum에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

May 27, 2026 Private analytics via zero-trust aggregation Security, Privacy and Abuse Prevention에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

Microsoft Research blog에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

Data Formulator 0.7: AI-powered data analytics for enterprise data에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

Extending Human Intelligence Through AI에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.