Classroom Final Exam: An Instructor-Tested Reasoning Benchmark

We introduce \CFE{} (\textbf{C}lassroom \textbf{F}inal \textbf{E}xam), a multimodal benchmark for evaluating the reasoning capabilities of large language models across more than 20 STEM domains. \CFE{} is curated from repeatedly used, authentic university homework and exam problems, together with reference solutions provided by course instructors. \CFE{} presents a significant challenge even for frontier models: the newly released Gemini-3.1-pro-preview achieves an overall accuracy of 59.69\%, while the second-best model, Gemini-3-flash-preview, reaches 55.46\%, leaving considerable room for improvement. Beyond leaderboard results, we perform a diagnostic analysis by decomposing reference solutions into reasoning flows. We find that although frontier models can often answer intermediate sub-questions correctly, they struggle to reliably derive and maintain correct intermediate states throughout multi-step solutions. We further observe that model-generated solutions typically have more reasoning steps than those provided by the instructor, indicating suboptimal step efficiency and a higher risk of error accumulation. The data and code are available at https://github.com/Analogy-AI/CFE_Bench.

Modeling Distinct Human Interaction in Web Agents

Despite rapid progress in autonomous web agents, human involvement remains essential for shaping preferences and correcting agent behavior as tasks unfold. However, current agentic systems lack a principled understanding of when and why humans intervene, often proceeding autonomously past critical decision points or requesting unnecessary confirmation. In this work, we introduce the task of modeling human intervention to support collaborative web task execution. We collect CowCorpus, a dataset of 400 real-user web navigation trajectories containing over 4,200 interleaved human and agent actions. We identify four distinct patterns of user interaction with agents -- hands-off supervision, hands-on oversight, collaborative task-solving, and full user takeover. Leveraging these insights, we train language models (LMs) to anticipate when users are likely to intervene based on their interaction styles, yielding a 61.4-63.4% improvement in intervention prediction accuracy over base LMs. Finally, we deploy these intervention-aware models in live web navigation agents and evaluate them in a user study, finding a 26.5% increase in user-rated agent usefulness. Together, our results show structured modeling of human intervention leads to more adaptive, collaborative agents.

Learning Personalized Agents from Human Feedback

Modern AI agents are powerful but often fail to align with the idiosyncratic, evolving preferences of individual users. Prior approaches typically rely on static datasets, either training implicit preference models on interaction history or encoding user profiles in external memory. However, these approaches struggle with new users and with preferences that change over time. We introduce Personalized Agents from Human Feedback (PAHF), a framework for continual personalization in which agents learn online from live interaction using explicit per-user memory. PAHF operationalizes a three-step loop: (1) seeking pre-action clarification to resolve ambiguity, (2) grounding actions in preferences retrieved from memory, and (3) integrating post-action feedback to update memory when preferences drift. To evaluate this capability, we develop a four-phase protocol and two benchmarks in embodied manipulation and online shopping. These benchmarks quantify an agent's ability to learn initial preferences from scratch and subsequently adapt to persona shifts. Our theoretical analysis and empirical results show that integrating explicit memory with dual feedback channels is critical: PAHF learns substantially faster and consistently outperforms both no-memory and single-channel baselines, reducing initial personalization error and enabling rapid adaptation to preference shifts.

Are Open-Weight LLMs Ready for Social Media Moderation? A Comparative Study on Bluesky

As internet access expands, so does exposure to harmful content, increasing the need for effective moderation. Research has demonstrated that large language models (LLMs) can be effectively utilized for social media moderation tasks, including harmful content detection. While proprietary LLMs have been shown to zero-shot outperform traditional machine learning models, the out-of-the-box capability of open-weight LLMs remains an open question. Motivated by recent developments of reasoning LLMs, we evaluate seven state-of-the-art models: four proprietary and three open-weight. Testing with real-world posts on Bluesky, moderation decisions by Bluesky Moderation Service, and annotations by two authors, we find a considerable degree of overlap between the sensitivity (81%--97%) and specificity (91%--100%) of the open-weight LLMs and those (72%--98%, and 93%--99%) of the proprietary ones. Additionally, our analysis reveals that specificity exceeds sensitivity for rudeness detection, but the opposite holds for intolerance and threats. Lastly, we identify inter-rater agreement across human moderators and the LLMs, highlighting considerations for deploying LLMs in both platform-scale and personalized moderation contexts. These findings show open-weight LLMs can support privacy-preserving moderation on consumer-grade hardware and suggest new directions for designing moderation systems that balance community values with individual user preferences.

The Llama 4 Herd: Architecture, Training, Evaluation, and Deployment Notes

This document consolidates publicly reported technical details about Metas Llama 4 model family. It summarizes (i) released variants (Scout and Maverick) and the broader herd context including the previewed Behemoth teacher model, (ii) architectural characteristics beyond a high-level MoE description covering routed/shared-expert structure, early-fusion multimodality, and long-context design elements reported for Scout (iRoPE and length generalization strategies), (iii) training disclosures spanning pre-training, mid-training for long-context extension, and post-training methodology (lightweight SFT, online RL, and lightweight DPO) as described in release materials, (iv) developer-reported benchmark results for both base and instruction-tuned checkpoints, and (v) practical deployment constraints observed across major serving environments, including provider-specific context limits and quantization packaging. The manuscript also summarizes licensing obligations relevant to redistribution and derivative naming, and reviews publicly described safeguards and evaluation practices. The goal is to provide a compact technical reference for researchers and practitioners who need precise, source-backed facts about Llama 4.

The Tool Decathlon: Benchmarking Language Agents for Diverse, Realistic, and Long-Horizon Task Execution

Real-world language agents must handle complex, multi-step workflows across diverse Apps. For instance, an agent may manage emails by coordinating with calendars and file systems, or monitor a production database to detect anomalies and generate reports following an operating manual. However, existing language agent benchmarks often focus on narrow domains or simplified tasks that lack the diversity, realism, and long-horizon complexity required to evaluate agents' real-world performance. To address this gap, we introduce the Tool Decathlon (dubbed as Toolathlon), a benchmark for language agents offering diverse Apps and tools, realistic environment setup, and reliable execution-based evaluation. Toolathlon spans 32 software applications and 604 tools, ranging from everyday platforms such as Google Calendar and Notion to professional ones like WooCommerce, Kubernetes, and BigQuery. Most of the tools are based on a high-quality set of Model Context Protocol (MCP) servers that we may have revised or implemented ourselves. Unlike prior works, which primarily ensure functional realism but offer limited environment state diversity, we provide realistic initial environment states from real software, such as Canvas courses with dozens of students or real financial spreadsheets. This benchmark includes 108 manually sourced or crafted tasks in total, requiring interacting with multiple Apps over around 20 turns on average to complete. Each task is strictly verifiable through dedicated evaluation scripts. Comprehensive evaluation of SOTA models highlights their significant shortcomings: the best-performing model, Claude-4.5-Sonnet, achieves only a 38.6% success rate with 20.2 tool calling turns on average, while the top open-weights model DeepSeek-V3.2-Exp reaches 20.1%. We expect Toolathlon to drive the development of more capable language agents for real-world, long-horizon task execution.

EnvInjection: Environmental Prompt Injection Attack to Multi-modal Web Agents

Multi-modal large language model (MLLM)-based web agents interact with webpage environments by generating actions based on screenshots of the webpages. Environmental prompt injection attacks manipulate the environment to induce the web agent to perform a specific, attacker-chosen action--referred to as the target action. However, existing attacks suffer from limited effectiveness or stealthiness, or are impractical in real-world settings. In this work, we propose EnvInjection, a new attack that addresses these limitations. Our attack adds a perturbation to the raw pixel values of the rendered webpage, which can be implemented by modifying the webpage's source code. After these perturbed pixels are mapped into a screenshot, the perturbation induces the web agent to perform the target action. We formulate the task of finding the perturbation as an optimization problem. A key challenge in solving this problem is that the mapping between raw pixel values and screenshot is non-differentiable, making it difficult to backpropagate gradients to the perturbation. To overcome this, we train a neural network to approximate the mapping and apply projected gradient descent to solve the reformulated optimization problem. Extensive evaluation on multiple webpage datasets shows that EnvInjection is highly effective and significantly outperforms existing baselines.

CowPilot: A Framework for Autonomous and Human-Agent Collaborative Web Navigation

While much work on web agents emphasizes the promise of autonomously performing tasks on behalf of users, in reality, agents often fall short on complex tasks in real-world contexts and modeling user preference. This presents an opportunity for humans to collaborate with the agent and leverage the agent's capabilities effectively. We propose CowPilot, a framework supporting autonomous as well as human-agent collaborative web navigation, and evaluation across task success and task efficiency. CowPilot reduces the number of steps humans need to perform by allowing agents to propose next steps, while users are able to pause, reject, or take alternative actions. During execution, users can interleave their actions with the agent by overriding suggestions or resuming agent control when needed. We conducted case studies on five common websites and found that the human-agent collaborative mode achieves the highest success rate of 95% while requiring humans to perform only 15.2% of the total steps. Even with human interventions during task execution, the agent successfully drives up to half of task success on its own. CowPilot can serve as a useful tool for data collection and agent evaluation across websites, which we believe will enable research in how users and agents can work together. Video demonstrations are available at https://oaishi.github.io/cowpilot.html

TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks

We interact with computers on an everyday basis, be it in everyday life or work, and many aspects of work can be done entirely with access to a computer and the Internet. At the same time, thanks to improvements in large language models (LLMs), there has also been a rapid development in AI agents that interact with and affect change in their surrounding environments. But how performant are AI agents at helping to accelerate or even autonomously perform work-related tasks? The answer to this question has important implications for both industry looking to adopt AI into their workflows, and for economic policy to understand the effects that adoption of AI may have on the labor market. To measure the progress of these LLM agents' performance on performing real-world professional tasks, in this paper, we introduce TheAgentCompany, an extensible benchmark for evaluating AI agents that interact with the world in similar ways to those of a digital worker: by browsing the Web, writing code, running programs, and communicating with other coworkers. We build a self-contained environment with internal web sites and data that mimics a small software company environment, and create a variety of tasks that may be performed by workers in such a company. We test baseline agents powered by both closed API-based and open-weights language models (LMs), and find that with the most competitive agent, 24% of the tasks can be completed autonomously. This paints a nuanced picture on task automation with LM agents -- in a setting simulating a real workplace, a good portion of simpler tasks could be solved autonomously, but more difficult long-horizon tasks are still beyond the reach of current systems.

Beyond Browsing: API-Based Web Agents

Web browsers are a portal to the internet, where much of human activity is undertaken. Thus, there has been significant research work in AI agents that interact with the internet through web browsing. However, there is also another interface designed specifically for machine interaction with online content: application programming interfaces (APIs). In this paper we ask -- what if we were to take tasks traditionally tackled by browsing agents, and give AI agents access to APIs? To do so, we propose two varieties of agents: (1) an API-calling agent that attempts to perform online tasks through APIs only, similar to traditional coding agents, and (2) a Hybrid Agent that can interact with online data through both web browsing and APIs. In experiments on WebArena, a widely-used and realistic benchmark for web navigation tasks, we find that API-based agents outperform web browsing agents. Hybrid Agents out-perform both others nearly uniformly across tasks, resulting in a more than 20.0% absolute improvement over web browsing alone, achieving a success rate of 35.8%, achiving the SOTA performance among task-agnostic agents. These results strongly suggest that when APIs are available, they present an attractive alternative to relying on web browsing alone.