Search for: All records

Understanding and reasoning about code semantics is essential for enhancing code LLMs' abilities to solve real-world software engineering (SE) tasks. Although several code reasoning benchmarks exist, most rely on synthetic datasets or educational coding problems and focus on coarse-grained reasoning tasks such as input/output prediction, limiting their effectiveness in evaluating LLMs in practical SE contexts. To bridge this gap, we propose CodeSense, the first benchmark that makes available a spectrum of fine-grained code reasoning tasks concerned with the software engineering of real-world code. We collected Python, C and Java software projects from real-world repositories. We executed tests from these repositories, collected their execution traces, and constructed a ground truth dataset for fine-grained semantic reasoning tasks. We then performed comprehensive evaluations on state-of-the-art LLMs. Our results show a clear performance gap for the models to handle fine-grained reasoning tasks. Although prompting techniques such as chain-of-thought and in-context learning helped, the lack of code semantics in LLMs fundamentally limit models' capabilities of code reasoning. Besides dataset, benchmark and evaluation, our work produced an execution tracing framework and tool set that make it easy to collect ground truth for fine-grained SE reasoning tasks, offering a strong basis for future benchmark construction and model post training. Our code and data are located at \url{https://codesense-bench.github.io/}. 

Modality fusion is a cornerstone of multimodal learning, enabling information integration from diverse data sources. However, vanilla fusion methods are limited by (1) inability to account for heterogeneous interactions between modalities and (2) lack of interpretability in uncovering the multimodal interactions inherent in the data. To this end, we propose I2MoE (Interpretable Multimodal Interaction-aware Mixture of Experts), an end-to-end MoE framework designed to enhance modality fusion by explicitly modeling diverse multimodal interactions, as well as providing interpretation on a local and global level. First, I2MoE utilizes different interaction experts with weakly supervised interaction losses to learn multimodal interactions in a data-driven way. Second, I2MoE deploys a reweighting model that assigns importance scores for the output of each interaction expert, which offers sample-level and dataset-level interpretation. Extensive evaluation of medical and general multimodal datasets shows that I2MoE is flexible enough to be combined with different fusion techniques, consistently improves task performance, and provides interpretation across various real-world scenarios. 

The Fermilab E906/SeaQuest collaboration performed measurements of the Drell-Yan process using 120 GeV proton beams bombarding liquid hydrogen and liquid deuterium targets. A combined analysis of all collected data was performed to obtain the final results for the ${\sigma }_{pd}/2{\sigma }_{pp}$ Drell-Yan cross section ratio covering the kinematic region of $0.13<x<0.45$ . The $x$ dependencies of $\overline{d}\left(x\right)/\overline{u}\left(x\right)$ and $\overline{d}\left(x\right)-\overline{u}\left(x\right)$ are extracted from these cross section ratios. It is found that $\overline{d}\left(x\right)$ is greater than $\overline{u}\left(x\right)$ over the entire measured $x$ range, with improved statistical accuracy compared to previous measurements. The new results on $\overline{d}\left(x\right)/\overline{u}\left(x\right)$ and $\overline{d}\left(x\right)-\overline{u}\left(x\right)$ are compared to various parton distribution functions and theoretical calculations. 

Euphemisms have not received much attention in natural language processing, despite being an important element of polite and figurative language. Euphemisms prove to be a difficult topic, not only because they are subject to language change, but also because humans may not agree on what is a euphemism and what is not. Nevertheless, the first step to tackling the issue is to collect and analyze examples of euphemisms. We present a corpus of potentially euphemistic terms (PETs) along with example texts from the GloWbE corpus. Additionally, we present a subcorpus of texts where these PETs are not being used euphemistically, which may be useful for future applications. We also discuss the results of multiple analyses run on the corpus. Firstly, we find that sentiment analysis on the euphemistic texts supports that PETs generally decrease negative and offensive sentiment. Secondly, we observe cases of disagreement in an annotation task, where humans are asked to label PETs as euphemistic or not in a subset of our corpus text examples. We attribute the disagreement to a variety of potential reasons, including if the PET was a commonly accepted term (CAT). 

We explore linguistic features that contribute to sarcasm detection. The linguistic features that we investigate are a combination of text and word complexity, stylistic and psychological features. We experiment with sarcastic tweets with and without context. The results of our experiments indicate that contextual information is crucial for sarcasm prediction. One important observation is that sarcastic tweets are typically incongruent with their context in terms of sentiment or emotional load.