--Automated crash reporting systems generate large volumes of duplicate reports, overwhelming issue-tracking systems and increasing developer workload. Traditional stack trace-based deduplication methods--relying on string similarity, rule-based heuristics, or deep learning (DL) models--often fail to capture the contextual and structural relationships within stack traces. We propose dedupT, a transformer-based approach that models stack traces holistically rather than as isolated frames. Extensive experiments on real-world datasets show that dedupT outperforms existing DL and traditional methods (e.g., sequence alignment and information retrieval techniques) in both duplicate ranking and unique crash detection, significantly reducing manual triage effort. On four public datasets, dedupT improves Mean Reciprocal Rank (MRR) often by over 15% compared to the best DL baseline and up to 9% over traditional methods while achieving higher Receiver Operating Characteristic Area Under the Curve (ROC-AUC) in detecting unique crash reports. Our work advances the integration of modern natural language processing (NLP) techniques into software engineering, providing an effective solution for stack trace-based crash deduplication. Software issues are generally reported through (1) human-submitted reports and (2) automated crash reports. Human-reported issues typically include textual descriptions detailing the issue, expected and observed behavior, and may include attachments such as images or videos. In contrast, automated crash reports are generated by crash reporting tools (e.g., Sentry However, these automated systems often overwhelm ITS platforms by generating numerous duplicate crash reports for the same issue, requiring developers to manually review and triage them, which is a time-consuming process. For instance, Mozilla Firefox received 2.2 million issues in the first week of 2016, the majority being duplicates [1], while 72% of crash reports in the IntelliJ Platform were found to be duplicates [2]. In such scenarios, grouping similar crashes together is essential, a process known as crash deduplication . Unlike human-written reports with detailed descriptions, automated crash reports primarily contain technical data like stack traces and crash dumps. Figure 1: Example of a Java stack trace. Figure 1: Example of C++ stack trace.

We present a mechanistic case study of a format-dependent reasoning failure in Llama-3.1-8B-Instruct, where the model incorrectly judges "9.11" as larger than "9.8" in chat or Q&A formats, but answers correctly in simple format. Through systematic intervention, we discover transformers implement even/odd attention head specialization: even indexed heads handle numerical comparison, while odd heads serve incompatible functions. The bug requires exactly 8 even heads at Layer 10 for perfect repair. Any combination of 8+ even heads succeeds, while 7 or fewer completely fails, revealing sharp computational thresholds with perfect redundancy among the 16 even heads. SAE analysis reveals the mechanism: format representations separate (10% feature overlap at Layer 7), then re-entangle with different weightings (80% feature overlap at Layer 10), with specific features showing 1.5x amplification in failing formats. We achieve perfect repair using only 25% of attention heads and identify a 60% pattern replacement threshold, demonstrating that apparent full-module requirements hide sophisticated substructure with implications for interpretability and efficiency. All of our code is available at https://github.com/gussand/surgeon.

Humor is a broad and complex form of communication that remains challenging for machines. Despite its broadness, most existing research on computational humor traditionally focused on modeling a specific type of humor. In this work, we wish to understand whether competence on one or more specific humor tasks confers any ability to transfer to novel, unseen types; in other words, is this fragmentation inevitable? This question is especially timely as new humor types continuously emerge in online and social media contexts (e.g., memes, anti-humor, AI fails). If Large Language Models (LLMs) are to keep up with this evolving landscape, they must be able to generalize across humor types by capturing deeper, transferable mechanisms. To investigate this, we conduct a series of transfer learning experiments across four datasets, representing different humor tasks. We train LLMs under varied diversity settings (1-3 datasets in training, testing on a novel task). Experiments reveal that models are capable of some transfer, and can reach up to 75% accuracy on unseen datasets; training on diverse sources improves transferability (1.88-4.05%) with minimal-to-no drop in in-domain performance. Further analysis suggests relations between humor types, with Dad Jokes surprisingly emerging as the best enabler of transfer (but is difficult to transfer to). We release data and code.

Analog subcircuit identification is a core task in analog design, essential for simulation, sizing, and layout. Traditional methods often require extensive human expertise, rule-based encoding, or large labeled datasets. To address these challenges, we propose GENIE-ASI, the first training-free, large language model (LLM)-based methodology for analog subcircuit identification. GENIE-ASI operates in two phases: it first uses in-context learning to derive natural language instructions from a few demonstration examples, then translates these into executable Python code to identify subcircuits in unseen SPICE netlists. In addition, to evaluate LLM-based approaches systematically, we introduce a new benchmark composed of operational amplifier netlists (op-amps) that cover a wide range of subcircuit variants. Experimental results on the proposed benchmark show that GENIE-ASI matches rule-based performance on simple structures (F1-score = 1.0), remains competitive on moderate abstractions (F1-score = 0.81), and shows potential even on complex subcircuits (F1-score = 0.31). These findings demonstrate that LLMs can serve as adaptable, general-purpose tools in analog design automation, opening new research directions for foundation model applications in analog design automation.

Recent progress in large language models (LLMs) has shown strong potential for multimodal reasoning beyond natural language. In this work, we explore the use of a fine-tuned Vision-Language Model (VLM), based on LLaMA 3.2, for classifying neutrino interactions from pixelated detector images in high-energy physics (HEP) experiments. We benchmark its performance against an established CNN baseline used in experiments like NOvA and DUNE, evaluating metrics such as classification accuracy, precision, recall, and AUC-ROC. Our results show that the VLM not only matches or exceeds CNN performance but also enables richer reasoning and better integration of auxiliary textual or semantic context. These findings suggest that VLMs offer a promising general-purpose backbone for event classification in HEP, paving the way for multimodal approaches in experimental neutrino physics.

Existing long-context benchmarks for Large Language Models (LLMs) focus on evaluating comprehension of long inputs, while overlooking the evaluation of long reasoning abilities. To address this gap, we introduce LongReasonArena, a benchmark specifically designed to assess the long reasoning capabilities of LLMs. Our tasks require models to solve problems by executing multi-step algorithms that reflect key aspects of long reasoning, such as retrieval and backtracking. By controlling the inputs, the required reasoning length can be arbitrarily scaled, reaching up to 1 million tokens of reasoning for the most challenging tasks. Extensive evaluation results demonstrate that LongReasonArena presents a significant challenge for both open-source and proprietary LLMs. For instance, Deepseek-R1 achieves only 7.5% accuracy on our task. Further analysis also reveals that the accuracy exhibits a linear decline with respect to the logarithm of the expected number of reasoning steps. Our code and data is available at https://github.com/LongReasonArena/LongReasonArena.

Event Extraction (EE) involves automatically identifying and extracting structured information about events from unstructured text, including triggers, event types, and arguments. Traditional discriminative models demonstrate high precision but often exhibit limited recall, particularly for nuanced or infrequent events. Conversely, generative approaches leveraging Large Language Models (LLMs) provide higher semantic flexibility and recall but suffer from hallucinations and inconsistent predictions. To address these challenges, we propose Agreement-based Reflective Inference System (ARIS), a hybrid approach combining a Self Mixture of Agents with a discriminative sequence tagger. ARIS explicitly leverages structured model consensus, confidence-based filtering, and an LLM reflective inference module to reliably resolve ambiguities and enhance overall event prediction quality. We further investigate decomposed instruction fine-tuning for enhanced LLM event extraction understanding. Experiments demonstrate our approach outperforms existing state-of-the-art event extraction methods across three benchmark datasets.

--Large Language Models (LLMs) excel in language tasks but are prone to hallucinations and outdated knowledge. However, in complex domains involving multiple, lengthy, or conflicting documents, traditional RAG suffers from information overload and inefficient synthesis, leading to inaccurate and untrustworthy answers. T o address this, we propose CASC (Context-Adaptive Synthesis and Compression), a novel framework that intelligently processes retrieved contexts. CASC introduces a Context Analyzer & Synthesizer (CAS) module, powered by a fine-tuned smaller LLM, which performs key information extraction, cross-document consistency checking and conflict resolution, and question-oriented structured synthesis. This process transforms raw, scattered information into a highly condensed, structured, and semantically rich context, significantly reducing the token count and cognitive load for the final Reader LLM. We evaluate CASC on SciDocs-QA, a new challenging multi-document question answering dataset designed for complex scientific domains with inherent redundancies and conflicts. Our extensive experiments demonstrate that CASC consistently outperforms strong baselines. Large Language Models (LLMs) have demonstrated remarkable capabilities in understanding, generating, and processing human language across a wide array of tasks [1]. However, despite their impressive fluency and reasoning abilities, LLMs inherently suffer from several critical limitations, including the propensity for "hallucinations" (generating factually incorrect or nonsensical information) and a knowledge cut-off date, rendering them incapable of accessing the most current information [2].

Agentic AI is advancing rapidly, yet truly autonomous medical-imaging triage, where a system decides when to stop, escalate, or defer under real constraints, remains relatively underexplored. To address this gap, we introduce AT-CXR, an uncertainty-aware agent for chest X-rays. The system estimates per-case confidence and distributional fit, then follows a stepwise policy to issue an automated decision or abstain with a suggested label for human intervention. We evaluate two router designs that share the same inputs and actions: a deterministic rule-based router and an LLM-decided router. Across five-fold evaluation on a balanced subset of NIH ChestX-ray14 dataset, both variants outperform strong zero-shot vision-language models and state-of-the-art supervised classifiers, achieving higher full-coverage accuracy and superior selective-prediction performance, evidenced by a lower area under the risk-coverage curve (AURC) and a lower error rate at high coverage, while operating with lower latency that meets practical clinical constraints. The two routers provide complementary operating points, enabling deployments to prioritize maximal throughput or maximal accuracy. Our code is available at https://github.com/XLIAaron/uncertainty-aware-cxr-agent.

With the widespread use of large language models (LLMs), understanding their potential failure modes during user interactions is essential. In practice, users often pose multiple questions in a single conversation with LLMs. Therefore, in this study, we propose Group Query Attack, a technique that simulates this scenario by presenting groups of queries to LLMs simultaneously. We investigate how the accumulated context from consecutive prompts influences the outputs of LLMs. Specifically, we observe that Group Query Attack significantly degrades the performance of models fine-tuned on specific tasks. Moreover, we demonstrate that Group Query Attack induces a risk of triggering potential backdoors of LLMs. Besides, Group Query Attack is also effective in tasks involving reasoning, such as mathematical reasoning and code generation for pre-trained and aligned models.