Ambient AI scribes draft clinical notes under the reassurance that a clinician signs every note. We audited three commercial AI scribes on the same 142 consultations: 565 notes from recorded UK primary-care and US ambulatory encounters plus authored scenarios. Twelve discovery passes proposed 13,678 candidate errors; the 5,898 clearing an importance filter went to an adversarial panel of two models from different families, each told to refute what it could, and 618 survived. One note in three (31.3% [27.0, 35.6]) carries a verified failure, concentrated in allergy and medication information, invented patient identity, and history written up as examination on telephone consultations that can contain none. No product was given a patient record; setting aside the two classes a record would have prefilled, invented identity and dates, the rate is 24.8% [20.8, 29.0]. One failure mode did not fit our scheme, drawn from published scribe-error taxonomies: a treatment the clinician retracts, recorded as delivered care. Two clinicians adjudicated blind, disjoint samples: a physician author upheld 20 of 21 findings (95.2% [77.3, 99.2]) and an independent clinician, not an author, 12 of 12 ([75.8, 100]); both judged every sampled refusal genuine. A failure rate depends on the instrument as much as the scribes. With model, evidence and settings fixed, the review instruction alone moves the share of candidates verified from 9.3% to 79.0%, and the reviewing family moves it too: alone at that instruction the gentler flags 54.8% of notes against 27.8%. Between 28% and 97% of sampled notes carry a failure depending on the standard. Published audits disagree among themselves by a margin instrument differences alone can produce: omission is 54-86% of their errors against our 23.1%. We release all 618 findings with transcript-side evidence, every prompt and model version, and the re-runnable pipeline.
Sebastian Fox, Luke Markham, Ryan Lail et al.· 0 citations
Ambient AI scribes draft clinical notes, and published audits find their dominant error is omission: information the encounter established that the note fails to record. The standard check is an LLM judge: a second model reads the note against the transcript and flags problems. We ask whether judges detect omissions. Public corpora cannot supply the answer key: their clinician reference notes and transcripts are materially discrepant. Our benchmark has 500 single-error note pairs from audited fact sheets, 298 with a named fact certainly absent and 202 added-or-altered controls. Across eight judge designs, paired discrimination (the flawed note below its clean twin, 0.5 a coin flip) reads 0.79-0.94 on added or altered content and 0.50-0.63 on omissions. On single notes, no design flags omissions reliably more often than perfect notes. Wording changes, voting and GEPA prompt optimisation move the operating point without creating usable detection. Restructuring the task recovers it: list the facts the transcript establishes, then check the note for each. Two methods reach it independently and trade off: a per-fact pipeline, and a GEPA-evolved prompt doing the same in one call. The pipeline's flags name the missing fact and its severity at 2.7% false alarms. The single call detects more (36.9% against 24.6%, p=0.002) at 6.2% false alarms and a tenth of the cost per note. A physician author validated 70 items and, where the two routes disagree, sided with the pipeline on 10 of 10 (p=0.002). A second clinician, not an author, graded the severity rubric blind and agrees to within a grade. On real vendor notes from a companion census no benchmark threshold transfers, but the re-calibrated single call detects more than the best of the eight at half its false-alarm rate. Omissions whose fact is restated elsewhere defeat both routes. We release the benchmark, prompts and judgements.
Sebastian Fox, Luke Markham, Ryan Lail et al.· 0 citations
Computational mental health screening using multimodal speech and text has shown great promise. However, existing models often assume all clinical speech protocols carry equivalent evidentiary validity. In reality, heterogeneous protocols, from free interviews to fixed reading tasks, support fundamentally different evidence. Forcing uniform reasoning flattens these boundaries, causing models to hallucinate symptoms from irrelevant text or overclaim support. Even advanced long chain-of-thought LLMs fail to resolve this issue, as free-form reasoning can exacerbate boundary violations. To address this, we reformulate multimodal screening as an evidence-bounded reasoning problem. We introduce the Evidence Package Benchmark, integrating 1,870 packages across six heterogeneous sources with explicit modality masks and evidence permissions. We further propose EviBound, a protocol-aware evidence control framework. Unlike direct LLM prompting, EviBound uses a profile-aware planner to restrict reasoning scope, orchestrates evidence tools via five-way acoustic consensus, and enforces a boundary critic to suppress unsupported claims. Empirical results show EviBound achieves a held-out test Depression AUROC of 0.8658, exceeding the strongest direct omni-modal baseline by +0.0811 AUROC while maintaining zero claim violations. Our work moves beyond unconstrained accuracy toward evidence-consistent, protocol-aware systems for safer clinical NLP research.
Chengyuan Gao, Jiang Wu, Tao Lu et al.· 0 citations
Retrieval-augmented generation systems can precompute and store key-value caches of retrieved documents to avoid re-encoding context at every query. Quantizing these caches further reduces storage, but no prior work asks whether compression damages faithfulness, whether responses remain grounded in the retrieved evidence. Faithfulness and accuracy are not equivalent: a model can produce a correct answer that is no longer supported by the context it was given. We evaluate Qwen2.5-7B-Instruct under INT8 and INT4 quantization on RGB and HotpotQA, measuring both accuracy and faithfulness with a hallucination detector, NLI entailment, and an LLM judge. INT8 is near-lossless across both metrics. INT4 reduces accuracy and, more critically, even among answers that remain factually correct, over 90% of faithfulness changes are negative, i.e., accuracy metrics are blind to this regression. The harm grows under noisy retrieval and with more retrieved chunks. Faithfulness must be audited before compressed caches are deployed.
Atta Ul Asad, Ahsan Bilal, Muhammad Ali et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
We study self-modeling: an LLM's ability to answer questions about its own behavior. We focus on verifiable behavioral questions, such as whether a prompt edit would change the model's final answer. To measure this capability, we introduce a benchmark that tests diverse types of self-modeling questions. Current models show non-trivial but limited self-modeling skill, and make systematic mistakes on simple counterfactual questions about their own behavior. To improve self-modeling skill, we develop a scalable synthetic-data pipeline that produces self-modeling training data, and show that reinforcement-learning can improve aggregate self-modeling skill across three open-source model families with some transfer to held-out tasks. These gains, however, do not seem to constitute introspection consistently: improved self-modeling may not arise from privileged access to the model's internal decision process.
Siqi Zeng, Andre N. Assis, Rowan Wang· 0 citations
We present CogEvol, a family of models trained specifically for Learning Environment Generation: turning a course brief into a finished learning artifact (structured-JSON slides or self-contained interactive HTML pages) in a single pass. Across 220k production requests, CogEvol completes a slide in a median of 17 seconds and an interactive page in 59, replacing minutes-long multi-turn agent scaffolding. Reliability is enforced rather than hoped for: a production-grounded data pipeline turns real failures into 53,687 verified SFT samples, and a hybrid rule-plus-VLM reward drives GRPO-based RL, hardened after we caught and fixed a reward-hacking episode that produced visually convincing but unplayable games. CogEvol-27B scores 83.7 on slide quality and 63.7 on a 500-case interactive-HTML benchmark with 26.9x fewer parameters than flagship coding models, and, in collaboration with the OpenMAIC team, serves their live production traffic. CogEvol-4B is released openly under the Apache 2.0 license at https://github.com/CogEvol/CogEvol-4B; external flagships are measured on the same suites under the identical harness. Scaffold editing cuts interactive-page generation cost by a further ~76%, and the full stack runs on domestic Ascend accelerators at application-level parity with A800 GPUs, lowering the unit cost of AI-native education at scale.
Shangqing Tu, Daniel Zhang-Li, Yucheng Wang et al.· 0 citations
LLMs can imitate how people write, which raises concerns about impersonation, trust, and detection in social settings. These concerns are especially important for adolescents, who use generative AI frequently but may struggle to recognize it. We introduce \textit{DoppelBot}, a cooperative social deduction game designed to study how young people detect and respond to AI impersonation. Through studies with middle schoolers, we investigate whether a DoppelBot prompts reflection on privacy and impersonation, how repeated exposure affects AI-detection accuracy as agents become more personalized, and which strategies students use to identify AI doppelg\"angers. We find that students' detection accuracy improves over time, driven by a shift from relying on linguistic cues to leveraging shared social and contextual signals. Students also demonstrated an understanding of AI limitations such as embodiment and reflected on broader issues such as data privacy. To support future research, we release an anonymized dataset of game transcripts and voting behavior.
Dan Schumacher, Pragathi Durga Rajarajan, Haven Kotara et al.· 0 citations
We present a family of retrieval methods for Polish statutory law built on document surrogates: language-model annotations attached to statutory articles at index time. Three designs occupy different points on the cost-quality frontier. ASCR is a surrogate cascade with reranking; ASCR-H fuses a dense list into that cascade; and DTF replaces both language-model stages with three lexical and dense retrievers, weighted reciprocal rank fusion, and a deterministic re-scoring prior, using no model call before generation. We evaluate all three against fourteen lexical, dense, fused and ablated baselines plus four controls, on 300 questions from the 2024 and 2025 Polish bar and legal counsel entrance examinations (264 with their reference article in the corpus), over 82,508 articles from 1,133 acts. On paired McNemar tests, ASCR-H places the reference provision at rank one significantly more often than every other non-oracle configuration except one of its own ablations (eighteen of twenty comparisons significant in its favour at p < 0.005), reaching 72.3% against 61.7% for BM25 and 52.3% for dense retrieval. The advantage is concentrated at the head and does not survive depth: it is significant at cutoffs of one and five, disappears by ten, and by twenty DTF leads on point estimate (86.0% versus 84.5%) at one ninth the latency and less than half the cost. Ablation attributes 27.6 points of rank-one accuracy to the reranking stage alone. We further report that the ranking advantage does not extend to citation accuracy, where DTF matches the oracle ceiling, and three negative results on lemmatisation, pseudo-relevance feedback and query rewriting. Surrogate annotation covers 27.0% of the corpus but every reference provision in the benchmark, an asymmetry we disclose and discuss. Benchmark, per-question outputs and paired significance tests are publicly available.
Travel itinerary generation requires balancing strict spatio-temporal constraints with human preferences. Existing LLM-based planners mainly rely on structured attributes and pre- defined traveler personas, but real travel deci- sions are often shaped by reviews that reveal experiential factors such as comfort, safety, ser- vice quality, ambiance, crowding, and hidden risks absent from structured databases. Incor- porating such review information is therefore critical to realistic, user-centric itinerary gen- eration. We propose TRIPPULSE1, a multi- agent framework for review-grounded travel planning. Instead of relying on a monolithic planner (and face context and reasoning bot- tlenecks), TRIPPULSE2 decomposes itinerary generation into specialized agents (each op- erating over localized contexts) for accom- modations, transportation, meals, attractions, and events, coordinated through a global or- chestrator with scheduling mechanisms that enforce temporal and budget feasibility. We augment TRIPCRAFT with 100K+ real-world reviews and introduce Review-Grounded Per- sona Alignment (RGPA), an LLM-as-a-Judge metric for evaluating alignment with human- centric travel experiences. Experiments across multiple trip durations and diverse proprietary and open-source models show that TRIPPULSE maintains strong constraint satisfaction while generating more personalized and experien- tially grounded itineraries.
Dynamic stance classification models how a reply responds to its direct parent message, rather than how a post relates to a fixed topic. Existing work has mainly studied this problem in text-only settings, while social media interactions increasingly rely on images, screenshots, memes, reaction images, and cross-modal references. We introduce MMDS-Bench, a diagnostic benchmark for multimodal dynamic stance classification in social media parent-reply interactions. MMDS-Bench contains 3,482 multimodal instances annotated with a seven-label dynamic stance taxonomy, together with an 800-instance diagnostic subset that requires structured reasoning over parent understanding, reply understanding, and stance-relation inference. We further annotate each instance with five challenge factors covering multimodal fusion, parent framing, non-literal expression, interaction reasoning, and label-boundary ambiguity. We evaluate 12 closed-source and open-source multimodal large language models and propose a reference-grounded LLM-judge protocol for assessing reasoning quality. Results show that current MLLMs still struggle with multimodal dynamic stance understanding, especially in cases that require relational inference beyond separate parent and reply comprehension.
Adapting large language models to user-specific preferences is often constrained by the cost of human annotation, making preference optimisation impractical in low-resource settings where preferences cannot be reliably labelled by LLMs themselves, e.g., due to cultural, subjective, or personalised contexts. In this paper, we investigate how language models encode preference information in their intermediate representations, finding that activations from chosen and rejected responses form distinct clusters across layers, even in pretrained models. Strikingly, this structure is strengthened by alignment on canonical datasets but erased when the target preferences differ from those the model was aligned on, suggesting aligned LLMs are poor judges for non-mainstream populations. Exploiting this structure, we propose training a lightweight linear probe on a few labelled preference pairs ($\leq$500) and using it to annotate large unlabelled datasets (50K+) for downstream preference optimisation. We systematically evaluate this approach across different datasets, preference optimisation methods and model scales and find that our method consistently outperforms direct training given the same annotation budget, and remains competitive against baselines trained on $50-100\times$ more labelled data in the majority of our settings. Code is available at https://github.com/alessioGalatolo/activ-pref-probe.
Electrocardiogram (ECG) interpretation requires knowledge of cardiology, electrophysiology, clinical diagnosis, ECG waveforms, signal acquisition, and instrumentation. Existing language-model benchmarks, however, primarily assess broad medical knowledge or interpretation of individual ECG signals and images rather than the broader contextual knowledge required for ECG interpretation. We developed ECGQuest, a literature-grounded resource for evaluating and fine-tuning ECG-specific language models. A GPT-4o-based pipeline generated questions from 23 ECG references and Computing in Cardiology proceedings from 2003-2025. The final dataset contains 10,904 unique True/False questions paired with their negated forms (21,808 Q&A pairs). We evaluated three commercial and 20 open-source language models on a held-out test set in a zero-shot setting. Five open-source models with 7-14B parameters were fine-tuned using Low-Rank Adaptation, with BERT and BiomedBERT included as supervised encoder baselines. Generalization was assessed on ECG-related subsets of MedMCQA and MedQA converted to binary True/False questions using official answer keys. Zero-shot accuracy on ECGQuest ranged from 49.5% to 74.4%, with GPT-5 performing best. General-purpose models outperformed medically specialized models, several models showed strong True/False bias, and encoder baselines performed near chance. Fine-tuning improved all open-source models by 6.5-14.1%. Fine-tuned DeepSeek-R1-Distill-Qwen-14B reached 76.3% accuracy, while a five-model voting ensemble reached 78.5%. On MedMCQA and MedQA, fine-tuning mainly benefited weaker or class-biased models and did not consistently improve strong base models. ECGQuest provides a reproducible benchmark for contextual ECG knowledge and shows that parameter-efficient fine-tuning can make smaller language models competitive with substantially larger commercial models.
Mohammadsina Hassannia, Matthew A. Reyna, Reza Sameni· 0 citations
With millions of users across the world, Julia has been used to conduct cutting-edge research and to design new drugs, jet engines, heat pumps, and more.
A new machine-learning framework aims to improve the success rate of computational protein design while moving away from results that reproduce sequences found in nature.
New MIT research could lead to better materials for a fossil-fuel-free process for making the chemical that's essential to fertilizer and other products.
