TimeRLM: Recursive Language Models Enable Precise Anomaly Localization in Long-Context Time-Series

Precise anomaly localization over long-context time series is a crucial task in monitoring applications across clinical care, industrial operations, financial services, and logistics, where brief evidence may hide inside long spans of high-frequency data. Time-Series Language Models (TSLMs) are able to ingest time series data and verbalize findings on anomalies in natural language; however, recent benchmarks report a decrease in retrieval performance at long contexts, mirroring failure modes in text, vision, and audio. In the text domain, Recursive Language Models (RLMs) can recover much of this lost performance by keeping context external to the large language model (LLM), allowing the model to query it through code. We present TimeRLM, an RLM formulation for time-series that sequentially manipulates the signal using code and vision capabilities. We further introduce AnomalyXL, a synthetic long-context anomaly localization benchmark with programmatically injected anomalies that require precise retrieval. We implement five different task categories and two variants: AnomalyXL-MCQ and AnomalyXL-Localize. TimeRLM outperforms every evaluated TSLM and single-pass baseline on four of the five AnomalyXL-Localize tasks, reaching 0.682 IoU on localization and 0.745 on classify-with-evidence, versus at most 0.329 and 0.072 across all baselines. We post-train TimeRLM using reinforcement learning. The resulting model further improves performance and requires approximately one-third as many agent interaction turns as its untrained base model to produce a final answer. On unseen real-world ECG, sleep and software observability recordings, the post-trained TimeRLM retains or improves performance, surpassing TSLMs despite being trained exclusively on synthetic data. Our findings suggest recursive interaction with time-series is an effective approach for long-horizon retrieval.

Patient-Level Multimodal Question Answering from Multi-Site Auscultation Recordings

Auscultation is a vital diagnostic tool, yet its utility is often limited by subjective interpretation. While general-purpose Audio-Language Models (ALMs) excel in general domains, they struggle with the nuances of physiological signals. We propose a framework that aligns multi-site auscultation recordings directly with a frozen Large Language Model (LLM) embedding space via gated cross-attention. By leveraging the LLM's latent world knowledge, our approach moves beyond isolated classification toward holistic, patient-level assessment. On the CaReSound benchmark, our model achieves a state-of-the-art 0.865 F1-macro and 0.952 BERTScore. We demonstrate that lightweight, domain-specific encoders rival large-scale ALMs and that multi-site aggregation provides spatial redundancy that mitigates temporal truncation. This alignment of medical acoustics with text foundations offers a scalable path for bridging signal processing and clinical assessment.

Surrogate-Based Prevalence Measurement for Large-Scale A/B Testing

Online media platforms often need to measure how frequently users are exposed to specific content attributes in order to evaluate trade-offs in A/B experiments. A direct approach is to sample content, label it using a high-quality rubric (e.g., an expert-reviewed LLM prompt), and estimate impression-weighted prevalence. However, repeatedly running such labeling for every experiment arm and segment is too costly and slow to serve as a default measurement at scale. We present a scalable \emph{surrogate-based prevalence measurement} framework that decouples expensive labeling from per-experiment evaluation. The framework calibrates a surrogate signal to reference labels offline and then uses only impression logs to estimate prevalence for arbitrary experiment arms and segments. We instantiate this framework using \emph{score bucketing} as the surrogate: we discretize a model score into buckets, estimate bucket-level prevalences from an offline labeled sample, and combine these calibrated bucket level prevalences with the bucket distribution of impressions in each arm to obtain fast, log-based estimates. Across multiple large-scale A/B tests, we validate that the surrogate estimates closely match the reference estimates for both arm-level prevalence and treatment--control deltas. This enables scalable, low-latency prevalence measurement in experimentation without requiring per-experiment labeling jobs.

Decision Quality Evaluation Framework at Pinterest

Online platforms require robust systems to enforce content safety policies at scale. A critical component of these systems is the ability to evaluate the quality of moderation decisions made by both human agents and Large Language Models (LLMs). However, this evaluation is challenging due to the inherent trade-offs between cost, scale, and trustworthiness, along with the complexity of evolving policies. To address this, we present a comprehensive Decision Quality Evaluation Framework developed and deployed at Pinterest. The framework is centered on a high-trust Golden Set (GDS) curated by subject matter experts (SMEs), which serves as a ground truth benchmark. We introduce an automated intelligent sampling pipeline that uses propensity scores to efficiently expand dataset coverage. We demonstrate the framework's practical application in several key areas: benchmarking the cost-performance trade-offs of various LLM agents, establishing a rigorous methodology for data-driven prompt optimization, managing complex policy evolution, and ensuring the integrity of policy content prevalence metrics via continuous validation. The framework enables a shift from subjective assessments to a data-driven and quantitative practice for managing content safety systems.

TS-Haystack: A Multi-Scale Retrieval Benchmark for Time Series Language Models

Time Series Language Models (TSLMs) are emerging as unified models for reasoning over continuous signals in natural language. However, long-context retrieval remains a major limitation: existing models are typically trained and evaluated on short sequences, while real-world time-series sensor streams can span millions of datapoints. This mismatch requires precise temporal localization under strict computational constraints, a regime that is not captured by current benchmarks. We introduce TS-Haystack, a long-context temporal retrieval benchmark comprising ten task types across four categories: direct retrieval, temporal reasoning, multi-step reasoning and contextual anomaly. The benchmark uses controlled needle insertion by embedding short activity bouts into longer longitudinal accelerometer recordings, enabling systematic evaluation across context lengths ranging from seconds to 2 hours per sample. We hypothesize that existing TSLM time series encoders overlook temporal granularity as context length increases, creating a task-dependent effect: compression aids classification but impairs retrieval of localized events. Across multiple model and encoding strategies, we observe a consistent divergence between classification and retrieval behavior. Learned latent compression preserves or improves classification accuracy at compression ratios up to 176$\times$, but retrieval performance degrades with context length, incurring in the loss of temporally localized information. These results highlight the importance of architectural designs that decouple sequence length from computational complexity while preserving temporal fidelity.