arXiv subjects

Sina Mansouri

Publications and source records attributed to Sina Mansouri.

6 recordsLinked to original sources

arXiv2026

VeriSim: A Configurable Framework for Stress-Testing Medical AI Under Patient Communication Noise

Medical large language models are typically evaluated on idealized patient cases that do not reflect how real patients communicate. We introduce VeriSim, a patient simulation framework that injects controllable noise along six clinically grounded communication dimensions while substantially preserving each patient's medical record. Truth adherence is supported by a verifier that extracts atomic claims from each candidate utterance and judges them against a UMLS-grounded vector index built with BioLORD embeddings, using the retrieved atoms' structured clinical metadata (e.g., drug class, anatomical site, treats-condition relations) rather than surface-text similarity alone. Across seven open-weight LLMs, realistic noise reduces diagnostic accuracy by 15-25 percentage points and increases conversation length by 34-55%; the 7-8B models degrade 1.4x more than 70B+ models. A board-certified physician and a licensed nurse rate VeriSim's conversations highly on truth, realism, clinical utility, and noise fidelity (inter-annotator agreement >= 0.80 across all dimensions), and an LLM-as-judge closely tracks their ratings, enabling scalable evaluation. We release VeriSim as an open-source framework for rigorous stress-testing of clinical AI: https://github.com/mohitmarvania/VeriSim

cs.AI↗

arXiv2026

Does Playing it Safe Count as Faithfulness? Reassessing LVLM Hallucination Mitigation Methods

Recent inference-time hallucination mitigation methods for large vision-language models (LVLMs) report strong gains on hallucination benchmarks. However, it remains unclear whether lower hallucination scores reflect improved multimodal grounding or more conservative generation. We evaluate six mitigation methods across three LVLMs and four benchmarks, including hallucination-focused evaluation and the diverse capability benchmark MMStar. Our analysis reveals two consistent patterns. First, hallucination reduction is often coupled with reduced informativeness: methods that lower hallucination rates also reduce object recall, visual coverage, or response detailedness. Second, improvements on hallucination benchmarks do not reliably transfer to broader multimodal capabilities, with methods showing inconsistent or degraded performance on fine-grained perception and reasoning tasks. Our findings suggest that current evaluation protocols may overestimate progress by rewarding conservative generation. We argue that hallucination mitigation should be evaluated as a faithfulness--informativeness--capability trade-off rather than through hallucination scores alone.

cs.CV↗

arXiv2026

Knowing Bias, Doing Better: Mitigating Social Bias in LLMs via Know-Bias Neuron Enhancement

Large language models (LLMs) exhibit social biases that reinforce harmful stereotypes, limiting their safe deployment. Most existing debiasing methods adopt a suppressive paradigm by modifying parameters, prompts, or neurons associated with biased behavior; however, such approaches are often brittle, weakly generalizable, data-inefficient, and prone to degrading general capability. We propose \textbf{KnowBias}, a lightweight and conceptually distinct framework that mitigates bias by strengthening, rather than suppressing, neurons encoding bias-knowledge. KnowBias identifies neurons encoding bias knowledge using a small set of bias-knowledge questions via attribution-based analysis, and selectively enhances them at inference time. This design enables strong debiasing while preserving general capabilities, generalizes across bias types and demographics, and is highly data efficient, requiring only a handful of simple yes/no questions and no retraining. Experiments across multiple benchmarks and LLMs demonstrate consistent state-of-the-art debiasing performance with minimal utility degradation. Data and code are available at https://github.com/JP-25/KnowBias.

cs.AI↗

arXiv2026

Stability-Aware Feature Design for Robust Watermark Detection in Machine-Generated Text

The widespread adoption of large language models (LLMs) has intensified the demand for principled methods to distinguish human from machine-generated text. Watermarking provides a promising avenue, yet existing detectors exhibit sharp performance deterioration under multiple paraphrasing and when applied to shorter texts. We introduce Pattern Stability Score (PSS), a novel detection framework that leverages local statistical features and stability dynamics across paraphrased variants. Specifically, the proposed method combines global and local z-score features with higher-order statistics of run-length patterns, enriched by autocorrelation signals and stability scores computed over paraphrase depth. Numerical evaluations are performed on three benchmark datasets (PG-19, CNN/DailyMail, and WikiText) using multiple LLMs (Llama-3-8B, Qwen2-7B) and paraphrasers (Mistral-7B, Qwen2-7B, Gemma-7B), systematically stress-testing robustness under up to eight rounds of paraphrasing. Compared to prior z-score thresholding baselines and some state-of-the-art deep learning methods, our approach improves detection AUC (area under the receiver operating characteristic curve) by over 10-15 percentage points across different token lengths. Additionally, extensive cross-domain experiments demonstrate that a single universal classifier generalizes across different LLMs, paraphrasers, and text domains without retraining, maintaining above 87.8% AUC even when all components differ from training.

cs.CL↗

arXiv2026

Batch Prompting Suppresses Overthinking Reasoning Under Constraint: How Batch Prompting Suppresses Overthinking in Reasoning Models

Large Reasoning Models (LRMs) achieve strong performance through explicit chain-of-thought reasoning but suffer from \textit{overthinking}: generating excessive reasoning tokens even for trivial queries. {Beyond inflating cost, overthinking can be self-defeating: models enter recursive self-doubt loops that exhaust token budgets without producing an answer, causing API timeouts that directly hurt accuracy.} We present an empirical study showing that \textbf{batch prompting}, originally introduced for throughput optimization, effectively suppresses overthinking at inference time. Across 13 diverse benchmarks with DeepSeek-R1 and OpenAI-o1, batch prompting {reduces reasoning tokens by 76\% (2{,}950$\mapsto$710), on average, while preserving or improving accuracy}. Through behavioral analysis, we find that batching induces three beneficial effects: (1) it reduces per-query reasoning effort when multiple queries share a context; (2) it enables pattern induction, where models generalize from earlier examples to solve later ones; and (3) it suppresses hedging behavior (e.g., ``\texttt{wait,}'' ``\texttt{let me double-check}'') that signals metacognitive loops. We also show that explicit prompt constraints (``\texttt{Use no more than 100 tokens in thinking.}'') fail to reduce overthinking; models either ignore them or sacrifice accuracy. These findings reframe batch prompting as more than a cost optimization: it is a practical inference-time technique that improves efficiency and reliability without model modification.

cs.CL↗

arXiv2025

Sparse 3D Perception for Rose Harvesting Robots: A Two-Stage Approach Bridging Simulation and Real-World Applications

The global demand for medicinal plants, such as Damask roses, has surged with population growth, yet labor-intensive harvesting remains a bottleneck for scalability. To address this, we propose a novel 3D perception pipeline tailored for flower-harvesting robots, focusing on sparse 3D localization of rose centers. Our two-stage algorithm first performs 2D point-based detection on stereo images, followed by depth estimation using a lightweight deep neural network. To overcome the challenge of scarce real-world labeled data, we introduce a photorealistic synthetic dataset generated via Blender, simulating a dynamic rose farm environment with precise 3D annotations. This approach minimizes manual labeling costs while enabling robust model training. We evaluate two depth estimation paradigms: a traditional triangulation-based method and our proposed deep learning framework. Results demonstrate the superiority of our method, achieving an F1 score of 95.6% (synthetic) and 74.4% (real) in 2D detection, with a depth estimation error of 3% at a 2-meter range on synthetic data. The pipeline is optimized for computational efficiency, ensuring compatibility with resource-constrained robotic systems. By bridging the domain gap between synthetic and real-world data, this work advances agricultural automation for specialty crops, offering a scalable solution for precision harvesting.

cs.RO↗