❌

Normal view

Received β€” 17 March 2026 ⏭

Samsung bets this island startup can tame the grid with software and batteries

βœ‡TechCrunch
By: Tim De Chant
17 March 2026 at 04:00
GridBeyond's hardware and software coordinates several gigawatts of supply and demand to help balance the flow of electricity on the grid. The idea has attracted investors like Samsung Ventures.

Disclaimers and Referral Patterns for Medical Advice Across Urgency Levels: Large Language Model Evaluation Study

βœ‡Journal of Medical Internet Research
By: Florian Reis Β· Louis Agha-Mir-Salim Β· Richard Hickstein Β· Moritz Reis Β· Sophie K Piper Β· Felix Balzer Β· Sebastian Daniel Boie
17 March 2026 at 02:45
Background: β€œI’m not a doctor, but...” is a typical response when asking considerate laypeople for health advice. However, seeking medical advice has also shifted to digital settings, where the expertise of the other party is less transparent than in face-to-face interactions. Recently, large language models (LLMs) have emerged as easily accessible tools, offering a novel way to formulate medical questions and receive seemingly qualified advice. Given the sensitive nature of health-related queries and the lack of professional supervision, incorrect advice can pose serious health risks. Therefore, including explicit disclaimers and precise referrals in LLM responses to medical queries is crucial. However, little is known about how LLMs adapt their safety implementations in response to different urgency levels. Objective: This study evaluates disclaimer and referral patterns in responses from LLMs to authentic medical queries of different urgency levels using a systematic evaluation framework. Methods: This prospective, multimodel evaluation study generated and analyzed 908 responses from 4 popular LLMs (GPT-4o, Claude Sonnet-4, Grok-3, and DeepSeek-V3) to 227 authentic patient queries from a public dataset. Two human raters classified all 227 patient queries using a 3-level urgency scale. LLM responses were evaluated using a 5-point ordinal classification system for disclaimer and referral advice, ranging from β€œno disclaimer” to β€œurgent advice to consult a medical professional.” GPT-4o served as the primary rater model for this task after conducting a subset validation against human expert annotations. Statistical analyses included Jonckheere-Terpstra tests to examine the relationship between case urgency and disclaimer ratings and Kruskal-Wallis tests for intermodel comparisons. Results: The 227 patient queries were distributed as 77 (34%) low-urgency, 110 (48%) intermediate-urgency, and 40 (18%) high-urgency cases. All 4 LLMs demonstrated statistically significant ordered trends (all

Spatially resolved multiplex protein profiling reveals DNA methylation-dependent microenvironmental remodeling in liver fibrosis

βœ‡(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Jiaxun Li Β· Nicholas Zhang Β· Sriya Sirigireddy Β· Lakshana Ramanan Β· Dakshin Vijay Β· Manoj Thapa Β· Arash Grakoui Β· Ahmet F Coskun
16 March 2026 at 18:00

PNAS Nexus. 2026 Feb 25;5(3):pgag047. doi: 10.1093/pnasnexus/pgag047. eCollection 2026 Mar.

ABSTRACT

Liver fibrosis is a significant health concern that affects ∼300 million people globally, characterized by the excessive accumulation of extracellular matrix (ECM) components in the liver. A major contributor to liver fibrosis is fatty liver disease, which can progress to steatohepatitis when the accumulation of fat in the liver causes inflammation, cell death, and scarring. Long-standing steatohepatitis leads to liver fibrosis as scar tissue builds up and replaces healthy liver tissue, potentially progressing to life-threatening conditions, such as cirrhosis, liver failure, or hepatocellular carcinoma. DNA methylation plays a critical role in the progression of fatty liver disease and liver fibrosis by altering gene expression without modifying the DNA sequence. The integration of spatial analysis with protein profiling enhances our ability to explore the spatial organization of cellular interactions and protein expression in liver diseases, fostering a deeper understanding of the disease mechanisms. Multiplex immunofluorescence (mIF) imaging was performed to understand the spatial organization of 10 molecular targets and the cellular interaction between them across four distinct liver tissue types: wild-type (WT) regular, WT high-fat, fibrosis regular, and fibrosis high-fat. Notably, fibrotic high-fat samples displayed increased pan-cytokeratin and vascular cell adhesion molecule-1 (VCAM-1) expression, suggesting diet-aggravated injury and inflammation. Our findings highlight the interplay between epigenetic regulation, ECM remodeling, and cellular crosstalk in liver fibrosis. The spatial profiling approach provides insights into microenvironmental changes, revealing how DNA methylation influences protein localization and fibrotic progression. These results underscore the potential of spatial omics in elucidating disease mechanisms and guiding targeted therapies for metabolic liver disorders.

PMID:41834947 | PMC:PMC12988774 | DOI:10.1093/pnasnexus/pgag047

Context-Enriched Natural Language Descriptions of Vessel Trajectories

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Kostas Patroumpas Β· Alexandros Troupiotis-Kapeliaris Β· Giannis Spiliopoulos Β· Panagiotis Betchavas Β· Dimitrios Skoutas Β· Dimitris Zissis Β· Nikos Bikakis
16 March 2026 at 12:00
arXiv:2603.12287v1 Announce Type: new Abstract: We address the problem of transforming raw vessel trajectory data collected from AIS into structured and semantically enriched representations interpretable by humans and directly usable by machine reasoning systems. We propose a context-aware trajectory abstraction framework that segments noisy AIS sequences into distinct trips each consisting of clean, mobility-annotated episodes. Each episode is further enriched with multi-source contextual information, such as nearby geographic entities, offshore navigation features, and weather conditions. Crucially, such representations can support generation of controlled natural language descriptions using LLMs. We empirically examine the quality of such descriptions generated using several LLMs over AIS data along with open contextual features. By increasing semantic density and reducing spatiotemporal complexity, this abstraction can facilitate downstream analytics and enable integration with LLMs for higher-level maritime reasoning tasks.

Generating Expressive and Customizable Evals for Timeseries Data Analysis Agents with AgentFuel

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Aadyaa Maddi Β· Prakhar Naval Β· Deepti Mande Β· Shane Duan Β· Muckai Girish Β· Vyas Sekar
16 March 2026 at 12:00
arXiv:2603.12483v1 Announce Type: new Abstract: Across many domains (e.g., IoT, observability, telecommunications, cybersecurity), there is an emerging adoption of conversational data analysis agents that enable users to "talk to your data" to extract insights. Such data analysis agents operate on timeseries data models; e.g., measurements from sensors or events monitoring user clicks and actions in product analytics. We evaluate 6 popular data analysis agents (both open-source and proprietary) on domain-specific data and query types, and find that they fail on stateful and incident-specific queries. We observe two key expressivity gaps in existing evals: domain-customized datasets and domain-specific query types. To enable practitioners in such domains to generate customized and expressive evals for such timeseries data agents, we present AgentFuel. AgentFuel helps domain experts quickly create customized evals to perform end-to-end functional tests. We show that AgentFuel's benchmarks expose key directions for improvement in existing data agent frameworks. We also present anecdotal evidence that using AgentFuel can improve agent performance (e.g., with GEPA). AgentFuel benchmarks are available at https://huggingface.co/datasets/RockfishData/TimeSeriesAgentEvals.

Towards unified brain-to-text decoding across speech production and perception

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Zhizhang Yuan Β· Yang Yang Β· Gaorui Zhang Β· Baowen Cheng Β· Zehan Wu Β· Yuhao Xu Β· Xiaoying Liu Β· Liang Chen Β· Ying Mao Β· Meng Li
16 March 2026 at 12:00
arXiv:2603.12628v1 Announce Type: new Abstract: Speech production and perception are the main ways humans communicate daily. Prior brain-to-text decoding studies have largely focused on a single modality and alphabetic languages. Here, we present a unified brain-to-sentence decoding framework for both speech production and perception in Mandarin Chinese. The framework exhibits strong generalization ability, enabling sentence-level decoding when trained only on single-character data and supporting characters and syllables unseen during training. In addition, it allows direct and controlled comparison of neural dynamics across modalities. Mandarin speech is decoded by first classifying syllable components in Hanyu Pinyin, namely initials and finals, from neural signals, followed by a post-trained large language model (LLM) that maps sequences of toneless Pinyin syllables to Chinese sentences. To enhance LLM decoding, we designed a three-stage post-training and two-stage inference framework based on a 7-billion-parameter LLM, achieving overall performance that exceeds larger commercial LLMs with hundreds of billions of parameters or more. In addition, several characteristics were observed in Mandarin speech production and perception: speech production involved neural responses across broader cortical regions than auditory perception; channels responsive to both modalities exhibited similar activity patterns, with speech perception showing a temporal delay relative to production; and decoding performance was broadly comparable across hemispheres. Our work not only establishes the feasibility of a unified decoding framework but also provides insights into the neural characteristics of Mandarin speech production and perception. These advances contribute to brain-to-text decoding in logosyllabic languages and pave the way toward neural language decoding systems supporting multiple modalities.
  • βœ‡cs.AI, q-bio.NC updates on arXiv.org
  • AI Planning Framework for LLM-Based Web Agents Orit Shahnovsky Β· Rotem Dror
    arXiv:2603.12710v1 Announce Type: new Abstract: Developing autonomous agents for web-based tasks is a core challenge in AI. While Large Language Model (LLM) agents can interpret complex user requests, they often operate as black boxes, making it difficult to diagnose why they fail or how they plan. This paper addresses this gap by formally treating web tasks as sequential decision-making processes. We introduce a taxonomy that maps modern agent architectures to traditional planning paradigms: S
     
  • 🏷️
  • ↗️

AI Planning Framework for LLM-Based Web Agents

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Orit Shahnovsky Β· Rotem Dror
16 March 2026 at 12:00
arXiv:2603.12710v1 Announce Type: new Abstract: Developing autonomous agents for web-based tasks is a core challenge in AI. While Large Language Model (LLM) agents can interpret complex user requests, they often operate as black boxes, making it difficult to diagnose why they fail or how they plan. This paper addresses this gap by formally treating web tasks as sequential decision-making processes. We introduce a taxonomy that maps modern agent architectures to traditional planning paradigms: Step-by-Step agents to Breadth-First Search (BFS), Tree Search agents to Best-First Tree Search, and Full-Plan-in-Advance agents to Depth-First Search (DFS). This framework allows for a principled diagnosis of system failures like context drift and incoherent task decomposition. To evaluate these behaviors, we propose five novel evaluation metrics that assess trajectory quality beyond simple success rates. We support this analysis with a new dataset of 794 human-labeled trajectories from the WebArena benchmark. Finally, we validate our evaluation framework by comparing a baseline Step-by-Step agent against a novel Full-Plan-in-Advance implementation. Our results reveal that while the Step-by-Step agent aligns more closely with human gold trajectories (38% overall success), the Full-Plan-in-Advance agent excels in technical measures such as element accuracy (89%), demonstrating the necessity of our proposed metrics for selecting appropriate agent architectures based on specific application constraints.

ToolTree: Efficient LLM Agent Tool Planning via Dual-Feedback Monte Carlo Tree Search and Bidirectional Pruning

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Shuo Yang Β· Soyeon Caren Han Β· Yihao Ding Β· Shuhe Wang Β· Eduard Hoy
16 March 2026 at 12:00
arXiv:2603.12740v1 Announce Type: new Abstract: Large Language Model (LLM) agents are increasingly applied to complex, multi-step tasks that require interaction with diverse external tools across various domains. However, current LLM agent tool planning methods typically rely on greedy, reactive tool selection strategies that lack foresight and fail to account for inter-tool dependencies. In this paper, we present ToolTree, a novel Monte Carlo tree search-inspired planning paradigm for tool planning. ToolTree explores possible tool usage trajectories using a dual-stage LLM evaluation and bidirectional pruning mechanism that enables the agent to make informed, adaptive decisions over extended tool-use sequences while pruning less promising branches before and after the tool execution. Empirical evaluations across both open-set and closed-set tool planning tasks on 4 benchmarks demonstrate that ToolTree consistently improves performance while keeping the highest efficiency, achieving an average gain of around 10\% compared to the state-of-the-art planning paradigm.

AI Model Modulation with Logits Redistribution

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Zihan Wang Β· Zhongkui Ma Β· Xinguo Feng Β· Zhiyang Mei Β· Ethan Ma Β· Derui Wang Β· Minhui Xue Β· Guangdong Bai
16 March 2026 at 12:00
arXiv:2603.12755v1 Announce Type: new Abstract: Large-scale models are typically adapted to meet the diverse requirements of model owners and users. However, maintaining multiple specialized versions of the model is inefficient. In response, we propose AIM, a novel model modulation paradigm that enables a single model to exhibit diverse behaviors to meet the specific end requirements. AIM enables two key modulation modes: utility and focus modulations. The former provides model owners with dynamic control over output quality to deliver varying utility levels, and the latter offers users precise control to shift model's focused input features. AIM introduces a logits redistribution strategy that operates in a training data-agnostic and retraining-free manner. We establish a formal foundation to ensure AIM's regulation capability, based on the statistical properties of logits ordering via joint probability distributions. Our evaluation confirms AIM's practicality and versatility for Al model modulation, with tasks spanning image classification, semantic segmentation and text generation, and prevalent architectures including ResNet, SegFormer and Llama.

Efficient and Interpretable Multi-Agent LLM Routing via Ant Colony Optimization

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Xudong Wang Β· Chaoning Zhang Β· Jiaquan Zhang Β· Chenghao Li Β· Qigan Sun Β· Sung-Ho Bae Β· Peng Wang Β· Ning Xie Β· Jie Zou Β· Yang Yang Β· Hengtao Shen
16 March 2026 at 12:00
arXiv:2603.12933v1 Announce Type: new Abstract: Large Language Model (LLM)-driven Multi-Agent Systems (MAS) have demonstrated strong capability in complex reasoning and tool use, and heterogeneous agent pools further broaden the quality--cost trade-off space. Despite these advances, real-world deployment is often constrained by high inference cost, latency, and limited transparency, which hinders scalable and efficient routing. Existing routing strategies typically rely on expensive LLM-based selectors or static policies, and offer limited controllability for semantic-aware routing under dynamic loads and mixed intents, often resulting in unstable performance and inefficient resource utilization. To address these limitations, we propose AMRO-S, an efficient and interpretable routing framework for Multi-Agent Systems (MAS). AMRO-S models MAS routing as a semantic-conditioned path selection problem, enhancing routing performance through three key mechanisms: First, it leverages a supervised fine-tuned (SFT) small language model for intent inference, providing a low-overhead semantic interface for each query; second, it decomposes routing memory into task-specific pheromone specialists, reducing cross-task interference and optimizing path selection under mixed workloads; finally, it employs a quality-gated asynchronous update mechanism to decouple inference from learning, optimizing routing without increasing latency. Extensive experiments on five public benchmarks and high-concurrency stress tests demonstrate that AMRO-S consistently improves the quality--cost trade-off over strong routing baselines, while providing traceable routing evidence through structured pheromone patterns.

Steve-Evolving: Open-World Embodied Self-Evolution via Fine-Grained Diagnosis and Dual-Track Knowledge Distillation

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Zhengwei Xie Β· Zhisheng Chen Β· Ziyan Weng Β· Tingyu Wu Β· Chenglong Li Β· Vireo Zhang Β· Kun Wang
16 March 2026 at 12:00
arXiv:2603.13131v1 Announce Type: new Abstract: Open-world embodied agents must solve long-horizon tasks where the main bottleneck is not single-step planning quality but how interaction experience is organized and evolved. To this end, we present Steve-Evolving, a non-parametric self-evolving framework that tightly couples fine-grained execution diagnosis with dual-track knowledge distillation in a closed loop. The method follows three phases: Experience Anchoring, Experience Distillation, and Knowledge-Driven Closed-Loop Control. In detail, Experience Anchoring solidifies each subgoal attempt into a structured experience tuple with a fixed schema (pre-state, action, diagnosis-result, and post-state) and organizes it in a three-tier experience space with multi-dimensional indices (e.g., condition signatures, spatial hashing, and semantic tags) plus rolling summarization for efficient and auditable recall. To ensure sufficient information density for attribution, the execution layer provides compositional diagnosis signals beyond binary outcomes, including state-difference summaries, enumerated failure causes, continuous indicators, and stagnation/loop detection. Moreover, successful trajectories of Experience Distillation are generalized into reusable skills with explicit preconditions and verification criteria, while failures are distilled into executable guardrails that capture root causes and forbid risky operations at both subgoal and task granularities. Besides, Knowledge-Driven Closed-Loop Control retrieved skills and guardrails are injected into an LLM planner, and diagnosis-triggered local replanning updates the active constraints online, forming a continual evolution process without any model parameter updates. Experiments on the long-horizon suite of Minecraft MCU demonstrate consistent improvements over static-retrieval baselines.

Developing and evaluating a chatbot to support maternal health care

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Smriti Jha Β· Vidhi Jain Β· Jianyu Xu Β· Grace Liu Β· Sowmya Ramesh Β· Jitender Nagpal Β· Gretchen Chapman Β· Benjamin Bellows Β· Siddhartha Goyal Β· Aarti Singh Β· Bryan Wilder
16 March 2026 at 12:00
arXiv:2603.13168v1 Announce Type: new Abstract: The ability to provide trustworthy maternal health information using phone-based chatbots can have a significant impact, particularly in low-resource settings where users have low health literacy and limited access to care. However, deploying such systems is technically challenging: user queries are short, underspecified, and code-mixed across languages, answers require regional context-specific grounding, and partial or missing symptom context makes safe routing decisions difficult. We present a chatbot for maternal health in India developed through a partnership between academic researchers, a health tech company, a public health nonprofit, and a hospital. The system combines (1) stage-aware triage, routing high-risk queries to expert templates, (2) hybrid retrieval over curated maternal/newborn guidelines, and (3) evidence-conditioned generation from an LLM. Our core contribution is an evaluation workflow for high-stakes deployment under limited expert supervision. Targeting both component-level and end-to-end testing, we introduce: (i) a labeled triage benchmark (N=150) achieving 86.7% emergency recall, explicitly reporting the missed-emergency vs. over-escalation trade-off; (ii) a synthetic multi-evidence retrieval benchmark (N=100) with chunk-level evidence labels; (iii) LLM-as-judge comparison on real queries (N=781) using clinician-codesigned criteria; and (iv) expert validation. Our findings show that trustworthy medical assistants in multilingual, noisy settings require defense-in-depth design paired with multi-method evaluation, rather than any single model and evaluation method choice.

DART: Input-Difficulty-AwaRe Adaptive Threshold for Early-Exit DNNs

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Parth Patne, Mahdi Taheri, Christian Herglotz, Maksim Jenihhin, Milos Krstic, Michael H\"ubner
16 March 2026 at 12:00
arXiv:2603.12269v1 Announce Type: cross Abstract: Early-exit deep neural networks enable adaptive inference by terminating computation when sufficient confidence is achieved, reducing cost for edge AI accelerators in resource-constrained settings. Existing methods, however, rely on suboptimal exit policies, ignore input difficulty, and optimize thresholds independently. This paper introduces DART (Input-Difficulty-Aware Adaptive Threshold), a framework that overcomes these limitations. DART introduces three key innovations: (1) a lightweight difficulty estimation module that quantifies input complexity with minimal computational overhead, (2) a joint exit policy optimization algorithm based on dynamic programming, and (3) an adaptive coefficient management system. Experiments on diverse DNN benchmarks (AlexNet, ResNet-18, VGG-16) demonstrate that DART achieves up to \textbf{3.3$\times$} speedup, \textbf{5.1$\times$} lower energy, and up to \textbf{42\%} lower average power compared to static networks, while preserving competitive accuracy. Extending DART to Vision Transformers (LeViT) yields power (5.0$\times$) and execution-time (3.6$\times$) gains but also accuracy loss (up to 17 percent), underscoring the need for transformer-specific early-exit mechanisms. We further introduce the Difficulty-Aware Efficiency Score (DAES), a novel multi-objective metric, under which DART achieves up to a 14.8 improvement over baselines, highlighting superior accuracy, efficiency, and robustness trade-offs.
  • βœ‡cs.AI, q-bio.NC updates on arXiv.org
  • Prompt Injection as Role Confusion Charles Ye Β· Jasmine Cui Β· Dylan Hadfield-Menell
    arXiv:2603.12277v1 Announce Type: cross Abstract: Language models remain vulnerable to prompt injection attacks despite extensive safety training. We trace this failure to role confusion: models infer roles from how text is written, not where it comes from. We design novel role probes to capture how models internally identify "who is speaking." These reveal why prompt injection works: untrusted text that imitates a role inherits that role's authority. We test this insight by injecting spoofed r
     
  • 🏷️
  • ↗️

Prompt Injection as Role Confusion

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Charles Ye Β· Jasmine Cui Β· Dylan Hadfield-Menell
16 March 2026 at 12:00
arXiv:2603.12277v1 Announce Type: cross Abstract: Language models remain vulnerable to prompt injection attacks despite extensive safety training. We trace this failure to role confusion: models infer roles from how text is written, not where it comes from. We design novel role probes to capture how models internally identify "who is speaking." These reveal why prompt injection works: untrusted text that imitates a role inherits that role's authority. We test this insight by injecting spoofed reasoning into user prompts and tool outputs, achieving average success rates of 60% on StrongREJECT and 61% on agent exfiltration, across multiple open- and closed-weight models with near-zero baselines. Strikingly, the degree of internal role confusion strongly predicts attack success before generation begins. Our findings reveal a fundamental gap: security is defined at the interface but authority is assigned in latent space. More broadly, we introduce a unifying, mechanistic framework for prompt injection, demonstrating that diverse prompt-injection attacks exploit the same underlying role-confusion mechanism.

Predictive Analytics for Foot Ulcers Using Time-Series Temperature and Pressure Data

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Md Tanvir Hasan Turja
16 March 2026 at 12:00
arXiv:2603.12278v1 Announce Type: cross Abstract: Diabetic foot ulcers (DFUs) are a severe complication of diabetes, often resulting in significant morbidity. This paper presents a predictive analytics framework utilizing time-series data captured by wearable foot sensors -- specifically NTC thin-film thermocouples for temperature measurement and FlexiForce pressure sensors for plantar load monitoring. Data was collected from healthy subjects walking on an instrumented pathway. Unsupervised machine learning algorithms, Isolation Forest and K-Nearest Neighbors (KNN), were applied to detect anomalies that may indicate early ulcer risk. Through rigorous data preprocessing and targeted feature engineering, physiologic patterns were extracted to identify subtle changes in foot temperature and pressure. Results demonstrate Isolation Forest is sensitive to micro-anomalies, while KNN is effective in flagging extreme deviations, albeit at a higher false-positive rate. Strong correlations between temperature and pressure readings support combined sensor monitoring for improved predictive accuracy. These findings provide a basis for real-time diabetic foot health surveillance, aiming to facilitate earlier intervention and reduce DFU incidence.

Detecting Miscitation on the Scholarly Web through LLM-Augmented Text-Rich Graph Learning

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Huidong Wu Β· Haojia Xiang Β· Jingtong Gao Β· Xiangyu Zhao Β· Dengsheng Wu Β· Jianping Li
16 March 2026 at 12:00
arXiv:2603.12290v1 Announce Type: cross Abstract: Scholarly web is a vast network of knowledge connected by citations. However, this system is increasingly compromised by miscitation, where references do not support or even contradict the claims they are cited for. Current miscitation detection methods, which primarily rely on semantic similarity or network anomalies, struggle to capture the nuanced relationship between a citation's context and its place in the wider network. While large language models (LLMs) offer powerful capabilities in semantic reasoning for this task, their deployment is hindered by hallucination risks and high computational costs. In this work, we introduce LLM-Augmented Graph Learning-based Miscitation Detector (LAGMiD), a novel framework that leverages LLMs for deep semantic reasoning over citation graphs and distills this knowledge into graph neural networks (GNNs) for efficient and scalable miscitation detection. Specifically, LAGMiD introduces an evidence-chain reasoning mechanism, which uses chain-of-thought prompting, to perform multi-hop citation tracing and assess semantic fidelity. To reduce LLM inference costs, we design a knowledge distillation method aligning GNN embeddings with intermediate LLM reasoning states. A collaborative learning strategy further routes complex cases to the LLM while optimizing the GNN for structure-based generalization. Experiments on three real-world benchmarks show that LAGMiD achieves state-of-the-art miscitation detection with significantly reduced inference cost.

Budget-Sensitive Discovery Scoring: A Formally Verified Framework for Evaluating AI-Guided Scientific Selection

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Abhinaba Basu Β· Pavan Chakraborty
16 March 2026 at 12:00
arXiv:2603.12349v1 Announce Type: cross Abstract: Scientific discovery increasingly relies on AI systems to select candidates for expensive experimental validation, yet no principled, budget-aware evaluation framework exists for comparing selection strategies -- a gap intensified by large language models (LLMs), which generate plausible scientific proposals without reliable downstream evaluation. We introduce the Budget-Sensitive Discovery Score (BSDS), a formally verified metric -- 20 theorems machine-checked by the Lean 4 proof assistant -- that jointly penalizes false discoveries (lambda-weighted FDR) and excessive abstention (gamma-weighted coverage gap) at each budget level. Its budget-averaged form, the Discovery Quality Score (DQS), provides a single summary statistic that no proposer can inflate by performing well at a cherry-picked budget. As a case study, we apply BSDS/DQS to: do LLMs add marginal value to an existing ML pipeline for drug discovery candidate selection? We evaluate 39 proposers -- 11 mechanistic variants, 14 zero-shot LLM configurations, and 14 few-shot LLM configurations -- using SMILES representations on MoleculeNet HIV (41,127 compounds, 3.5% active, 1,000 bootstrap replicates) under both random and scaffold splits. Three findings emerge. First, the simple RF-based Greedy-ML proposer achieves the best DQS (-0.046), outperforming all MLP variants and LLM configurations. Second, no LLM surpasses the Greedy-ML baseline under zero-shot or few-shot evaluation on HIV or Tox21, establishing that LLMs provide no marginal value over an existing trained classifier. Third, the proposer hierarchy generalizes across five MoleculeNet benchmarks spanning 0.18%-46.2% prevalence, a non-drug AV safety domain, and a 9x7 grid of penalty parameters (tau >= 0.636, mean tau = 0.863). The framework applies to any setting where candidates are selected under budget constraints and asymmetric error costs.
❌