Customer-obsessed science
Research areas
-
November 16, 202611 min readNew open-source tool turns existing test suites into sets of formal-verification tasks, which catch buggy solutions that tests alone would miss.
-
November 11, 20265 min read
-
-
-
Featured news
-
NeurIPS 2026 Workshop on Generalization for Time Series in Tight Settings (TS-LIMITS)2026Large language models (LLMs) can adapt anomaly detection to heterogeneous time series, but recent LLM-based detectors keep the model in the scoring loop, coupling score production to repeated context construction and model inference. We introduce LLMDP-U, which uses an LLM as a detector compiler: from a statistical profile of a reference prefix and a numerical interface, it generates and validates a series-specific
-
ASE 20262026Traditional Infrastructure as Code misconfiguration remediation requires manual analysis, a labor-intensive approach that cannot meet the scale and speed demands of modern cloud deployments. We present IacFix, an agentic tool that combines LLM-based agents with formal methods to enable automated IaC security remediation from security findings to validated patches. IacFix introduces end-to-end remediation
-
RecSys 20262026Disambiguating broad user intent is an important challenge in e-commerce, as multiple possible meanings of ambiguous queries can result in completely different product selections. Such queries can be disambiguated in the context of the shopping session, as prior interactions provide crucial cues about users' underlying preferences. Existing query classification models ignore the cues from shopping sessions
-
RecSys 20262026A shared search-and-recommendation index has to score new items from their features alone, because search has no exploration slot in which to warm them up. On a public log carrying both surfaces over one catalog, 38.6% of query-search impressions in a held-out week show an item never previously shown or visited. On a narrower slice — engagements where the user, not the catalog, had no prior contact with
-
2026Uncertainty estimation (UE) enables LLM-powered systems to recognize when to abstain, yet existing research has predominantly focused on English. We present the first large-scale evaluation of UE methods across 22 languages, spanning high-, mid-, and low-resource settings. Using two human-curated Q&A datasets, we compare open and closed box UE methods (nine in total) across different model sizes and architectures
Collaborations
View allWhether you're a faculty member or student, there are number of ways you can engage with Amazon.
View all