Publications
Our teams aspire to make discoveries that impact everyone, and core to our approach is sharing our research and tools to fuel progress in the field.
Our teams aspire to make discoveries that impact everyone, and core to our approach is sharing our research and tools to fuel progress in the field.
Sort By
1 - 15 of 11377 publications
Preview abstract
We introduce AASE (Activation-based AI Safety Enforcement), a framework for post-perception safety monitoring in large language models. Unlike pre-perception approaches that analyze input or output text, AASE monitors the model's internal activation patterns—what the model "understands" rather than what text it processes or generates—enabling detection of safety-relevant states before harmful outputs are produced. The framework comprises three techniques: Activation Fingerprinting (AF) for harmful content detection, Agent Action Gating (AAG) for prompt injection defense, and Activation Policy Compliance (APC) for enterprise policy enforcement. We introduce paired contrastive training to isolate safety-relevant signals from confounding factors such as topic and style, addressing signal entanglement in polysemantic activations. Validation across 7 models from 3 architecture families shows strong class separation: Gemma-2-9B achieves AUC 1.00 with 7.2σ separation across all probes; AAG achieves AUC ≥0.88 across all models on the InjecAgent benchmark; APC achieves 0.97-1.00 AUC across three enterprise policies. Model size correlates with probe quality—Gemma-2-9B (7.2σ separation) outperforms Gemma-2-2B (4.3σ). All techniques survive INT4 quantization with minimal separation degradation. AASE is 9× faster than Llama Guard 3 (33ms vs 306ms) with higher TPR (88% vs 50%) at a tunable threshold that trades FPR for detection sensitivity, adding only 0.002ms probe overhead to existing inference.
View details
Abstain and Validate: A Dual-LLM Policy for Reducing Noise in Agentic Program Repair
José Cambronero
Michele Tufano
Renyao Wei
Grant Uy
Chin-Jung Liu
Shiying Pan
Satish Chandra
ICSE-SEIP '26: Proceedings of the IEEE/ACM 48th International Conference on Software Engineering: Software Engineering in Practice, ACM (Association for Computing Machinery), New York, NY, USA (2026), 118 – 129
Preview abstract
Agentic automated program repair (APR) is increasingly being used to tackle complex, repository-level bugs in both academic and industry settings. However, changes made by the agent still need to be reviewed by a human before committing them to the codebase to ensure the change correctly fixes the bug. Showing unlikely patches to developers can lead to substantial noise, wasting valuable developer time and decreasing trust in all automated code changes. We introduce two complementary LLM-based policies to reduce such noise: bug abstention and patch critic policies. Bug abstention allows us to exclude bugs that the agentic APR system is unlikely to fix and the patch critic rejects patches that are unlikely to be a good fix for the given bug. We evaluate both policies on three sets of bugs collected from Google’s codebase and agent trajectories generated by an agentic APR system in use at Google. We introduce filtered success rates, where we compute performance after removing trajectories rejected by our policies. On a set of 174 human-reported bugs, abstention and critic can raise filtered success rates up to 13 percentage points and 15 percentage points, respectively, and up to 29 percentage points in combination. For null pointer exceptions and sanitizer-reported bugs with machine-generated bug reports or known reproduction tests, we show the same critic policy can be used to improve average single-sample success rates.
View details
Preview abstract
This framework manages AI agents by establishing behavioral boundaries and a persistent identity. It uses a multi-layered stack, combining safety rules with brand guidelines, to shape an agent's reasoning. Features include authority decay to limit power if confidence drops and memory segmentation to prevent data tampering. Centralized oversight ensures these digital representatives remain aligned with company policies through continuous monitoring and testing.
View details
Expert evaluation of LLM world models: A high-Tc superconductivity case study
Haoyu Guo
Maria Tikhanovskaya
Paul Raccuglia
Alexey Vlaskin
Chris Co
Scott Ellsworth
Matthew Abraham
Lizzie Dorfman
Peter Armitage
Chunhan Feng
Antoine Georges
Olivier Gingras
Dominik Kiese
Steve Kivelson
Vadim Oganesyan
Brad Ramshaw
Subir Sachdev
Senthil Todadri
John Tranquada
Eun-Ah Kim
Proceedings of the National Academy of Sciences (2026)
Preview abstract
Large Language Models (LLMs) show great promise as a powerful tool for scientific literature exploration. However, their effectiveness in providing scientifically accurate and comprehensive answers to complex questions within specialized domains remains an active area of research. This work evaluates the performance of six different LLM-based systems for answering scientific literature questions, including commercially available closed models and a custom retrieval-augmented generation (RAG) system capable of retrieving images alongside text. We conduct a rigorous expert evaluation of the systems in the domain of high-temperature cuprate superconductors, a research area that involves material science, experimental physics, computation, and theoretical physics. We use an expert-curated database of 1726 scientific papers and a set of 67 expert-formulated questions. The evaluation employs a multi-faceted rubric assessing balanced perspectives, factual comprehensiveness, succinctness, evidentiary support, and image relevance. Our results demonstrate that RAG-based systems, powered by curated data and multimodal retrieval, outperform existing closed models across key metrics, particularly in providing comprehensive and well-supported answers, and in retrieving relevant visual information. This study provides valuable insights into designing and evaluating specialized scientific literature understanding systems, particularly with expert involvement, while also highlighting the importance of rich, domain-specific data in such systems.
View details
Preview abstract
Ideation, i. e., generating, capturing, organizing, and evaluating ideas, is foundational to creative practices; yet little is known about how blind and low vision ( BLV) creators engage in ideation. We conducted interviews with 20 BLV creators working in different domains including poetry, music production, tactile graphics, digital drawing, and mixed- media arts. Our findings reveal how participants established accessible creative workflows under social and institutional constraints, drew inspiration from diverse physical and digital sources, captured and organized ideas across text, audio, visual, and physical modalities, and evaluated their work through human and AI feedback. We discuss how ableist norms and accessibility breakdowns cascade across ideation pathways, progressively narrowing BLV individuals’ creative possibilities. Finally, we offer design considerations for building accessible ideation support tools that amplify BLV creators’ distinctive practices.
View details
Type-Aware Ranking of Urban Similarity from Aerial Imagery
Idan Kligvasser
Yotam Intrator
Yuval Desheh
Aviad Barzilai
Niv Efron
Ehud Rivlin
Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Workshops (2026), pp. 821-829
Preview abstract
Estimating and ranking cross-city similarity from aerial imagery is a fundamental challenge in remote sensing and geospatial representation learning. Urban environments differ widely in road layout, marking conventions, and infrastructure design, yet standard visual representations often struggle to disentangle these meaningful structural variations from superficial appearances. In this work, we propose a type-aware contrastive learning framework that measures urban similarity by explicitly modeling distinct infrastructure elements. Leveraging open-vocabulary retrieval, we construct a globally diverse dataset of road-related features, such as intersections, crosswalks, and bus lanes, and train a type-conditioned Vision Transformer that fuses visual features with CLIP-derived semantic embeddings. Crucially, we introduce an adaptive per-type contrastive loss that dynamically emphasizes infrastructure categories with high discriminative power while down-weighting less informative types. To quantify city-level similarity, we aggregate per-type cosine similarities via a lightweight classifier to generate a global city-to-city similarity matrix. Experiments demonstrate that this type-aware approach significantly improves clustering quality and successfully generalizes to unseen cities, establishing a scalable, interpretable foundation for comparative urban analysis.
View details
Preview abstract
Validating conversational artificial intelligence (AI) for regulated medical software applications may present challenges, as static test datasets and manual review may be limited in identifying emergent, conversational anomalies. A multi-agent AI system may be configured in a closed-loop for automated validation. The system can, for example, utilize an end user persona simulator agent to generate prompts for a target model and a domain /regulatory expert adjudicator agent to evaluate the target model’s responses against a configurable rubric. A meta-analysis agent can analyze anomalies to identify underlying vulnerabilities, which may then be used to programmatically synthesize new adversarial personas. This adaptive process can generate evidence to support regulatory compliance and continuous performance monitoring for medical software algorithms systems.
View details
Preview abstract
Generative AI’s humanlike qualities are driving its rapid adoption in professional domains. However, this anthropomorphic appeal raises concerns from HCI and responsible AI scholars about potential hazards and harms, such as overtrust in system outputs. To investigate how technology workers navigate these humanlike qualities and anticipate emergent harms, we conducted focus groups with 30 professionals across six job functions (ML engineering, product policy, UX research and design, product management, technology writing, and communications). Our findings reveal an unsettled knowledge environment surrounding humanlike generative AI, where workers’ varying perspectives illuminate a range of potential risks for individuals, knowledge work fields, and society. We argue that workers require comprehensive support, including clearer conceptions of “humanlikeness” to effectively mitigate these risks. To aid in mitigation strategies, we provide a conceptual map articulating the identified hazards and their connection to conflated notions of “humanlikeness.”
View details
Who Controls the Curriculum for AI? The Limits of Participatory Design for Educational AI
Learning Under Algorithmic Conditions, University of Minnesota Press (2026)
Preview abstract
Participatory design is a long-standing effort to shift control over technology design from technologists to users and communities impacted by technologies. For educational AI, this means involving students, families, teachers, and other stakeholders in shaping the design of AI systems. While promising, in this article, I situate the recent calls for participatory design of educational AI systems within a different historical tradition—that of contests over local control of educational curricula. I argue that approaches that attempt to steer the design and development of educational AI through participatory methods may inadvertently reproduce the history of political contestation of educational curricula, in ways that may privilege the most powerful communities, rather than those inequitably impacted. What might it look like to treat participatory AI design as a site for political contestation? How might these approaches avoid reproducing the same majoritarian tendencies that led to educational inequities in the first place?
View details
Preview abstract
Enterprise service centers, particularly in domains like People Operations, are critical hubs of organizational knowledge work. They face a persistent difficulty in disseminating the tacit, case-specific expertise of senior agents, which can lead to inconsistent service and slower onboarding for new hires. While existing Knowledge Management (KM) and Case-Based Reasoning (CBR) systems have improved the retrieval of historically similar cases, they inadvertently shift the cognitive burden of synthesizing this information to the time-constrained agent. This paper introduces the Dynamic Case Precedent (DCP) architecture, a novel socio-technical framework designed to address this gap. The DCP architecture moves beyond simple precedent recommendation to automated precedent synthesis. It achieves this by integrating a semantic retrieval model with the large-context reasoning capabilities of a generative Large Language Model (LLM). We propose a three-pillar framework—(1) Contextual Similarity Indexing, (2) Generative Insight Synthesis, and (3) Human-in-the-Loop Refinement. By analyzing multiple relevant historical cases to generate a concise summary of resolution patterns, the DCP architecture aims to reduce agent cognitive load, accelerate proficiency, and improve service consistency. This conceptual framework offers a new model for human-AI collaboration, framing the AI not as a mere information tool, but as an active partner in sensemaking.
View details
Phoenix: Rowhammer Attacks on DDR5 with Self-Correcting Synchronization
Michele Marazzi
Kaveh Razavi
Salman Qazi
Diego Meyer
Patrick Jattke
IEEE Security & Privacy (S&P) (2026)
Preview abstract
We introduce a new context-enriched time series forecasting benchmark TimesX. TimesX contains a wide selection of high-quality real-world time series and diverse textual contexts from an automated generating pipeline, which helps address three main issues of existing benchmarks: (1) poor generalization due to low data volume and data being synthetic, (2) restricted forms of context, and (3) an inability to mitigate data leakage. We conduct a thorough empirical study of current multimodal solutions on TimesX. Our results suggest that most multimodal solutions that work well on existing benchmarks may fail on TimesX. In contrast, simple ensemble methods that leverage the rich textual context can outperform strong unimodal baselines and other multimodal baselines.
** Below this is what was submitted to ITP. **
We create a real world multimodal time-series forecasting benchmark that encompasses diverse domains and regions. Each time-series is annotated by various kinds of contexts like metadata, date and holiday information, dynamic events related to the time-series. This is sufficiently more advanced than other available benchmarks which rely wither on static metadata alone or synthetic examples. This forms a test bed for multimodal forecasting. We also present some baseline results showing that ensembles of publicly available LLMs and time-series foundation models can demonstrate non-trivial performance on this bechmark.
View details
The Ontic-Epistemic Distinction: Implications for General Intelligence
Master's Thesis (2026) (to appear)
Preview abstract
The current pursuit of robust machine intelligence is largely predicated on a substrate independent, computational functionalist view of cognition, where sufficiently complex computational processing is expected to eventually yield generalized reasoning. This paper explores the ontological distinctions between these computational frameworks and biological cognition, specifically how these differences impact the capacity for semantic understanding. By analyzing phenomena such as the "reversal curse" where models fail to generalize the symmetry in identity relations (A=B implies B=A), and performance on novel reasoning benchmarks (e.g., ARC-AGI), this paper examines whether current model limitations are transient artifacts of scale or indicative of a distinct architectural category. Integrating Stevan Harnad’s “symbol grounding problem” with Evan Thompson’s biological model of “intrinsic normativity,” I investigate whether robust general intelligence might require sense-making: a process distinct from information processing, whereby an agent’s internal states are causally coupled with its environment via survival or system-wide stakes which grounds symbols in meaning. Current Large Language Models (LLMs) appear to lack this intrinsic normativity, and consequently may operate primarily as epistemic instruments rather than ontic agents. By introducing the concept of “ontic grounding”, this paper presents a potential framework for distinguishing between the simulation of reasoning and true understanding, which could have implications for AI safety and governance.
View details
Ten Insights from Other Domains That Inform Responsible AI Frameworks
Allison Woodruff
Angela McKay
Dunstan Allison-Hope
Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (2026), 104–115
Preview abstract
The rapid growth of AI systems is being accompanied by new guidelines, principles, standards, regulations, and best practices (hereafter “frameworks”) that seek to ensure the responsible design, development, deployment, and use of AI systems.
Our premise is that the substance, implementation, and evolution of these AI frameworks can be informed by the practical experience of pursuing similar desired outcomes in other relevant domains (e.g., content moderation, human rights, climate change). This will help ensure that mistakes are not repeated and more rapid progress is made.
We used a “repetition test” to generate the following ten insights from other domains. Insights passing the “repetition test” are those that experts with thousands of hours of practical experience often repeat when describing the best practices that have emerged from their domain.
AI frameworks can draw from these ten insights, rather than invent entirely new approaches.
View details
In Search of "Weird Corners": Diagnosing the Limits of Convergent AI in Professional Creative Practice
Mark Williams
2026
Preview abstract
Generative AI assistants typically employ convergent interaction paradigms to resolve ambiguity. While effective for technical tasks, this risks premature convergence in creative domains, constraining output variance. Evaluating a convergent AI probe with expert creatives (N=9) indicates an interactional paradox: structural linearity provides "ignition" utility for early ideation, but misaligns with organic workflows, often inducing "aesthetic sanitization" that standardizes individualized nuance. Prioritizing constructive friction over default agreement, the experts requested active, lateral collaborators. In response, we reframe output convergence as a "full-stack" UI challenge, advocating for Generative frameworks that operationalize the Double Diamond via fluid role-shifting and productive tension.
View details
