Weiyi Wang

Weiyi Wang, Xinchi Chen, Jingjing Gong et al.

Recent advances in agentic Large Language Models (LLMs) have positioned them as generalist planners capable of reasoning and acting across diverse tasks. However, existing agent benchmarks largely focus on symbolic or weakly grounded environments, leaving their performance in physics-constrained real-world domains underexplored. We introduce AstroReason-Bench, a comprehensive benchmark for evaluating agentic planning in Space Planning Problems (SPP), a family of high-stakes problems with heterogeneous objectives, strict physical constraints, and long-horizon decision-making. AstroReason-Bench integrates multiple scheduling regimes, including ground station communication and agile Earth observation, and provides a unified agent-oriented interaction protocol. Evaluating on a range of state-of-the-art open- and closed-source agentic LLM systems, we find that current agents substantially underperform specialized solvers, highlighting key limitations of generalist planning under realistic constraints. AstroReason-Bench offers a challenging and diagnostic testbed for future agentic research.

Gemini Team, Petko Georgiev, Ving Ian Lei et al. · deepmind, mila

In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content.

Gheorghe Comanici, Eric Bieber, Mike Schaekermann et al. · amazon-science, baidu

In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving.

Leo Linqian Gan, Jeffery Wu, Longyuan Ge et al.

Autonomous LLM agents face a critical security risk known as workflow hijacking, where attackers subtly alter tool and skill invocations. Existing defenses rely on host-internal telemetry (such as audit logs), which can be forged if the host OS is compromised. To solve this, we introduce ClawGuard, a passive, out-of-band monitor that audits LLM-agent workflows using electromagnetic (EM) emanations. Because distinct agent skills create unique hardware usage patterns (computation, DRAM, network blocking), they emit measurable, macroscopic EM envelopes. External software-defined radios (SDRs) capture these physical signals. Using a drift-aware pipeline with 320-dimensional features, ClawGuard converts RF streams into physical evidence. Evaluated on a 7.82TB RF corpus, ClawGuard achieved an AUC of 0.9945, detecting attacks with a 100% true-positive rate and a 1.16% false-positive rate. This proves passive EM sensing is a practical, forge-resistant physical check against compromised host software.

Henrique Schechter Vera, Sahil Dua, Biao Zhang et al.

We introduce EmbeddingGemma, a new lightweight, open text embedding model based on the Gemma 3 language model family. Our innovative training recipe strategically captures knowledge from larger models via encoder-decoder initialization and geometric embedding distillation. We improve model robustness and expressiveness with a spread-out regularizer, and ensure generalizability by merging checkpoints from varied, optimized mixtures. Evaluated on the Massive Text Embedding Benchmark (MTEB) across multilingual, English, and code domains, EmbeddingGemma (300M) achieves state-of-the-art results. Notably, it outperforms prior top models, both proprietary and open, with fewer than 500M parameters, and provides performance comparable to models double its size, offering an exceptional performance-to-cost ratio. Remarkably, this lead persists when quantizing model weights or truncating embedding outputs. This makes EmbeddingGemma particularly well-suited for low-latency and high-throughput use cases such as on-device applications. We provide ablation studies exploring our key design choices. We release EmbeddingGemma to the community to promote further research.

Weiyi Wang, Junwei Deng, Yuzheng Hu et al.

Data attribution methods, which quantify the influence of individual training data points on a machine learning model, have gained increasing popularity in data-centric applications in modern AI. Despite a recent surge of new methods developed in this space, the impact of hyperparameter tuning in these methods remains under-explored. In this work, we present the first large-scale empirical study to understand the hyperparameter sensitivity of common data attribution methods. Our results show that most methods are indeed sensitive to certain key hyperparameters. However, unlike typical machine learning algorithms -- whose hyperparameters can be tuned using computationally-cheap validation metrics -- evaluating data attribution performance often requires retraining models on subsets of training data, making such metrics prohibitively costly for hyperparameter tuning. This poses a critical open challenge for the practical application of data attribution methods. To address this challenge, we advocate for better theoretical understandings of hyperparameter behavior to inform efficient tuning strategies. As a case study, we provide a theoretical analysis of the regularization term that is critical in many variants of influence function methods. Building on this analysis, we propose a lightweight procedure for selecting the regularization value without model retraining, and validate its effectiveness across a range of standard data attribution benchmarks. Overall, our study identifies a fundamental yet overlooked challenge in the practical application of data attribution, and highlights the importance of careful discussion on hyperparameter selection in future method development.

Boqi Li, Siyuan Li, Weiyi Wang et al.

With the rapid progress of foundation models and robotics, vision-language navigation (VLN) has emerged as a key task for embodied agents with broad practical applications. We address VLN in continuous environments, a particularly challenging setting where an agent must jointly interpret natural language instructions, perceive its surroundings, and plan low-level actions. We propose a zero-shot framework that integrates a simplified yet effective waypoint predictor with a multimodal large language model (MLLM). The predictor operates on an abstract obstacle map, producing linearly reachable waypoints, which are incorporated into a dynamically updated topological graph with explicit visitation records. The graph and visitation information are encoded into the prompt, enabling reasoning over both spatial structure and exploration history to encourage exploration and equip MLLM with local path planning for error correction. Extensive experiments on R2R-CE and RxR-CE show that our method achieves state-of-the-art zero-shot performance, with success rates of 41% and 36%, respectively, outperforming prior state-of-the-art methods.

Ge Fan, Wei Zeng, Shan Sun et al.

In recent years, deep neural network is introduced in recommender systems to solve the collaborative filtering problem, which has achieved immense success on computer vision, speech recognition and natural language processing. On one hand, deep neural network can be used to model the auxiliary information in recommender systems. On the other hand, it is also capable of modeling nonlinear relationships between users and items. One advantage of deep neural network is that the performance of the algorithm can be easily enhanced by augmenting the depth of the neural network. However, two potential problems may emerge when the deep neural work is exploited to model relationships between users and items. The fundamental problem is that the complexity of the algorithm grows significantly with the increment in the depth of the neural network. The second one is that a deeper neural network may undermine the accuracy of the algorithm. In order to alleviate these problems, we propose a hybrid neural network that combines heterogeneous neural networks with different structures. The experimental results on real datasets reveal that our method is superior to the state-of-the-art methods in terms of the item ranking.