In this paper, we present an LLM-based code translation method and an associated tool called CoTran, that translates whole-programs from one high-level programming language to another. Current LLM-based code translation methods lack a training approach to ensure that the translated code reliably compiles or bears substantial functional equivalence to the input code. In our work, we train an LLM via reinforcement learning, by modifying the fine-tuning process to incorporate compiler feedback and symbolic execution (symexec)-based equivalence testing feedback that checks for functional equivalence between the input and output programs. The idea is to guide an LLM-in-training, via compiler and symexec-based testing feedback, by letting it know how far it is from producing perfect translations. We report on extensive experiments comparing CoTran with 14 other code translation tools that include human-written transpilers, LLM-based translation tools, and ChatGPT over a benchmark of more than 57,000 Java-Python equivalent pairs, and we show that CoTran outperforms them on relevant metrics such as compilation accuracy (CompAcc) and functional equivalence accuracy (FEqAcc). For example, our tool achieves 48.68% FEqAcc, 76.98% CompAcc for Python-to-Java translation, whereas the nearest competing tool (PLBART-base) only gets 38.26% and 75.77% resp. Also, built upon CodeT5, CoTran achieves +11.23%, +14.89% improvement on FEqAcc and +4.07%, +8.14% on CompAcc for Java-to-Python and Python-to-Java translation resp.
相關內容
Building scalable vision-language models to learn from diverse, multimodal data remains an open challenge. In this paper, we introduce an Efficient Vision-languagE foundation model, namely EVE, which is one unified multimodal Transformer pre-trained solely by one unified pre-training task. Specifically, EVE encodes both vision and language within a shared Transformer network integrated with modality-aware sparse Mixture-of-Experts (MoE) modules, which capture modality-specific information by selectively switching to different experts. To unify pre-training tasks of vision and language, EVE performs masked signal modeling on image-text pairs to reconstruct masked signals, i.e., image pixels and text tokens, given visible signals. This simple yet effective pre-training objective accelerates training by 3.5x compared to the model pre-trained with Image-Text Contrastive and Image-Text Matching losses. Owing to the combination of the unified architecture and pre-training task, EVE is easy to scale up, enabling better downstream performance with fewer resources and faster training speed. Despite its simplicity, EVE achieves state-of-the-art performance on various vision-language downstream tasks, including visual question answering, visual reasoning, and image-text retrieval.
In this paper we revisit the classical method of partitioning classification and study its convergence rate under relaxed conditions, both for observable (non-privatised) and for privatised data. Let the feature vector $X$ take values in $\mathbb{R}^d$ and denote its label by $Y$. Previous results on the partitioning classifier worked with the strong density assumption, which is restrictive, as we demonstrate through simple examples. We assume that the distribution of $X$ is a mixture of an absolutely continuous and a discrete distribution, such that the absolutely continuous component is concentrated to a $d_a$ dimensional subspace. Here, we study the problem under much milder assumptions: in addition to the standard Lipschitz and margin conditions, a novel characteristic of the absolutely continuous component is introduced, by which the exact convergence rate of the classification error probability is calculated, both for the binary and for the multi-label cases. Interestingly, this rate of convergence depends only on the intrinsic dimension $d_a$. The privacy constraints mean that the data $(X_1,Y_1), \dots ,(X_n,Y_n)$ cannot be directly observed, and the classifiers are functions of the randomised outcome of a suitable local differential privacy mechanism. The statistician is free to choose the form of this privacy mechanism, and here we add Laplace distributed noises to the discontinuations of all possible locations of the feature vector $X_i$ and to its label $Y_i$. Again, tight upper bounds on the rate of convergence of the classification error probability are derived, without the strong density assumption, such that this rate depends on $2\,d_a$.
In this work we present an approach for generating alternative text (or alt-text) descriptions for images shared on social media, specifically Twitter. More than just a special case of image captioning, alt-text is both more literally descriptive and context-specific. Also critically, images posted to Twitter are often accompanied by user-written text that despite not necessarily describing the image may provide useful context that if properly leveraged can be informative. We address this task with a multimodal model that conditions on both textual information from the associated social media post as well as visual signal from the image, and demonstrate that the utility of these two information sources stacks. We put forward a new dataset of 371k images paired with alt-text and tweets scraped from Twitter and evaluate on it across a variety of automated metrics as well as human evaluation. We show that our approach of conditioning on both tweet text and visual information significantly outperforms prior work, by more than 2x on BLEU@4.
In this paper, we consider the generation and utilization of helper data for physical unclonable functions (PUFs) that provide real-valued readout symbols. Compared to classical binary PUFs, more entropy can be extracted from each basic building block (PUF node), resulting in longer keys/fingerprints and/or a higher reliability. To this end, a coded modulation and signal shaping scheme that matches the (approximately) Gaussian distribution of the readout has to be employed. A new helper data scheme is proposed that works with any type of coded modulation/shaping scheme. Compared to the permutation scheme from the literature, less amount of helper data has to be generated and a higher reliability is achieved. Moreover, the recently proposed idea of a two-metric helper data scheme is generalized to coded modulation and a general S-metric scheme. It is shown how extra helper data can be generated to improve decodability. The proposed schemes are assessed by numerical simulations and by evaluation of measurement data. We compare multi-level codes using a new rate design strategy with bit-interleaved coded modulation and trellis shaping with a distribution matcher. By selecting a suitable design, the rate per PUF node that can be reliably extracted can be as high as 2~bit/node.
This paper proposes a framework for designing robust precoders for a multi-input single-output (MISO) system that performs integrated sensing and communication (ISAC) across multiple cells and users. We use Cramer-Rao-Bound (CRB) to measure the sensing performance and derive its expressions for two multi-cell scenarios, namely coordinated beamforming (CBF) and coordinated multi-point (CoMP). In the CBF scheme, a BS shares channel state information (CSI) and estimates target parameters using monostatic sensing. In contrast, a BS in the CoMP scheme shares the CSI and data, allowing bistatic sensing through inter-cell reflection. We consider both block-level (BL) and symbol-level (SL) precoding schemes for both the multi-cell scenarios that are robust to channel state estimation errors. The formulated optimization problems to minimize the CRB in estimating the parameters of a target and maximize the minimum communication signal-to-interference-plus-noise-ratio (SINR) while satisfying a given total transmit power budget are non-convex. We tackle the non-convexity using a combination of semidefinite relaxation (SDR) and alternating optimization (AO) techniques. Simulations suggest that neglecting the inter-cell reflection and communication links degrades the performance of an ISAC system. The CoMP scenario employing SL precoding performs the best, whereas the BL precoding applied in the CBF scenario produces relatively high estimation error for a given minimum SINR value.
This paper introduces the concept of Distributed Intelligent integrated Sensing and Communications (DISAC), which expands the capabilities of Integrated Sensing and Communications (ISAC) towards distributed architectures. Additionally, the DISAC framework integrates novel waveform design with new semantic and goal-oriented communication paradigms, enabling ISAC technologies to transition from traditional data fusion to the semantic composition of diverse sensed and shared information. This progress facilitates large-scale, energy-efficient support for high-precision spatial-temporal processing, optimizing ISAC resource utilization, and enabling effective multi-modal sensing performance. Addressing key challenges such as efficient data management and connect-compute resource utilization, 6G- DISAC stands to revolutionize applications in diverse sectors including transportation, healthcare, and industrial automation. Our study encapsulates the project vision, methodologies, and potential impact, marking a significant stride towards a more connected and intelligent world.
Debugging is famously one the hardest parts in programming. In this paper, we tackle the question: what does a debugging environment look like when we take interactive visualization as a central design principle? We introduce Anteater, an interactive visualization system for tracing and exploring the execution of Python programs. Existing systems often have visualization components built on top of an existing infrastructure. In contrast, Anteater's organization of trace data enables an intermediate representation which can be leveraged to automatically synthesize a variety of visualizations and interactions. These interactive visualizations help with tasks such as discovering important structures in the execution and understanding and debugging unexpected behaviors. To assess the utility of Anteater, we conducted a participant study where programmers completed tasks on their own python programs using Anteater. Finally, we discuss limitations and where further research is needed.
This article presents the affordances that Generative Artificial Intelligence can have in disinformation context, one of the major threats to our digitalized society. We present a research framework to generate customized agent-based social networks for disinformation simulations that would enable understanding and evaluation of the phenomena whilst discussing open challenges.
The problem of answering questions using knowledge from pre-trained language models (LMs) and knowledge graphs (KGs) presents two challenges: given a QA context (question and answer choice), methods need to (i) identify relevant knowledge from large KGs, and (ii) perform joint reasoning over the QA context and KG. In this work, we propose a new model, QA-GNN, which addresses the above challenges through two key innovations: (i) relevance scoring, where we use LMs to estimate the importance of KG nodes relative to the given QA context, and (ii) joint reasoning, where we connect the QA context and KG to form a joint graph, and mutually update their representations through graph neural networks. We evaluate QA-GNN on the CommonsenseQA and OpenBookQA datasets, and show its improvement over existing LM and LM+KG models, as well as its capability to perform interpretable and structured reasoning, e.g., correctly handling negation in questions.
In order to answer natural language questions over knowledge graphs, most processing pipelines involve entity and relation linking. Traditionally, entity linking and relation linking has been performed either as dependent sequential tasks or independent parallel tasks. In this paper, we propose a framework called "EARL", which performs entity linking and relation linking as a joint single task. EARL uses a graph connection based solution to the problem. We model the linking task as an instance of the Generalised Travelling Salesman Problem (GTSP) and use GTSP approximate algorithm solutions. We later develop EARL which uses a pair-wise graph-distance based solution to the problem.The system determines the best semantic connection between all keywords of the question by referring to a knowledge graph. This is achieved by exploiting the "connection density" between entity candidates and relation candidates. The "connection density" based solution performs at par with the approximate GTSP solution.We have empirically evaluated the framework on a dataset with 5000 questions. Our system surpasses state-of-the-art scores for entity linking task by reporting an accuracy of 0.65 to 0.40 from the next best entity linker.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191