Bridging Relevance and Reasoning: Rationale Distillation in Retrieval-Augmented Generation

The reranker and generator are two critical components in the Retrieval-Augmented Generation (i.e., RAG) pipeline, responsible for ranking relevant documents and generating responses. However, due to differences in pre-training data and objectives, there is an inevitable gap between the documents ranked as relevant by the reranker and those required by the generator to support answering the query. To address this gap, we propose RADIO, a novel and practical preference alignment framework with RAtionale DIstillatiOn. Specifically, We first propose a rationale extraction method that leverages the reasoning capabilities of Large Language Models (LLMs) to extract the rationales necessary for answering the query. Subsequently, a rationale-based alignment process is designed to rerank the documents based on the extracted rationales, and fine-tune the reranker to align the preferences. We conduct extensive experiments on two tasks across three datasets to demonstrate the effectiveness of our approach compared to baseline methods. Our code is released online to ease reproduction.

AltFS: Agency-light Feature Selection with Large Language Models in Deep Recommender Systems

Feature selection is crucial in recommender systems for improving model efficiency and predictive performance. Traditional methods rely on agency models, such as decision trees or neural networks, to estimate feature importance. However, this approach is inherently limited, as the agency models may fail to learn effectively in all scenarios due to suboptimal training conditions (e.g., feature collinearity, high-dimensional sparsity, and data insufficiency). In this paper, we propose AltFS, an Agency-light Feature Selection method for deep recommender systems. AltFS integrates semantic reasoning from Large Language Models (LLMs) with task-specific learning from agency models. Initially, LLMs will generate a semantic ranking of feature importance, which is then refined by an agency model, combining world knowledge with task-specific insights. Extensive experiments on three public datasets from real-world recommender platforms demonstrate the effectiveness of AltFS. Our code is publicly available for reproducibility.

MOSABench: Multi-Object Sentiment Analysis Benchmark for Evaluating Multimodal Large Language Models Understanding of Complex Image

Multimodal large language models (MLLMs) have shown remarkable progress in high-level semantic tasks such as visual question answering, image captioning, and emotion recognition. However, despite advancements, there remains a lack of standardized benchmarks for evaluating MLLMs performance in multi-object sentiment analysis, a key task in semantic understanding. To address this gap, we introduce MOSABench, a novel evaluation dataset designed specifically for multi-object sentiment analysis. MOSABench includes approximately 1,000 images with multiple objects, requiring MLLMs to independently assess the sentiment of each object, thereby reflecting real-world complexities. Key innovations in MOSABench include distance-based target annotation, post-processing for evaluation to standardize outputs, and an improved scoring mechanism. Our experiments reveal notable limitations in current MLLMs: while some models, like mPLUG-owl and Qwen-VL2, demonstrate effective attention to sentiment-relevant features, others exhibit scattered focus and performance declines, especially as the spatial distance between objects increases. This research underscores the need for MLLMs to enhance accuracy in complex, multi-object sentiment analysis tasks and establishes MOSABench as a foundational tool for advancing sentiment analysis capabilities in MLLMs.

Model Editing for LLMs4Code: How Far are We?

Large Language Models for Code (LLMs4Code) have been found to exhibit outstanding performance in the software engineering domain, especially the remarkable performance in coding tasks. However, even the most advanced LLMs4Code can inevitably contain incorrect or outdated code knowledge. Due to the high cost of training LLMs4Code, it is impractical to re-train the models for fixing these problematic code knowledge. Model editing is a new technical field for effectively and efficiently correcting erroneous knowledge in LLMs, where various model editing techniques and benchmarks have been proposed recently. Despite that, a comprehensive study that thoroughly compares and analyzes the performance of the state-of-the-art model editing techniques for adapting the knowledge within LLMs4Code across various code-related tasks is notably absent. To bridge this gap, we perform the first systematic study on applying state-of-the-art model editing approaches to repair the inaccuracy of LLMs4Code. To that end, we introduce a benchmark named CLMEEval, which consists of two datasets, i.e., CoNaLa-Edit (CNLE) with 21K+ code generation samples and CodeSearchNet-Edit (CSNE) with 16K+ code summarization samples. With the help of CLMEEval, we evaluate six advanced model editing techniques on three LLMs4Code: CodeLlama (7B), CodeQwen1.5 (7B), and Stable-Code (3B). Our findings include that the external memorization-based GRACE approach achieves the best knowledge editing effectiveness and specificity (the editing does not influence untargeted knowledge), while generalization (whether the editing can generalize to other semantically-identical inputs) is a universal challenge for existing techniques. Furthermore, building on in-depth case analysis, we introduce an enhanced version of GRACE called A-GRACE, which incorporates contrastive learning to better capture the semantics of the inputs.

Phase retrieval: Global convergence of gradient descent with optimal sample complexity

This paper addresses the phase retrieval problem, which aims to recover a signal vector $x$ from $m$ measurements $y_i=|\langle a_i,x^{\natural}\rangle|^2$, $i=1,\ldots,m$. A standard approach is to solve a nonconvex least squares problem using gradient descent with random initialization, which is known to work efficiently given a sufficient number of measurements. However, whether $O(n)$ measurements suffice for gradient descent to recover the ground truth efficiently has remained an open question. Prior work has established that $O(n\,{\rm poly}(\log n))$ measurements are sufficient. In this paper, we resolve this open problem by proving that $m=O(n)$ Gaussian random measurements are sufficient to guarantee, with high probability, that the objective function has a benign global landscape. This sample complexity is optimal because at least $\Omega(n)$ measurements are required for exact recovery. The landscape result allows us to further show that gradient descent with a constant step size converges to the ground truth from almost any initial point.

Learning Code Preference via Synthetic Evolution

Large Language Models (LLMs) have recently demonstrated remarkable coding capabilities. However, assessing code generation based on well-formed properties and aligning it with developer preferences remains challenging. In this paper, we explore two key questions under the new challenge of code preference learning: (i) How do we train models to predict meaningful preferences for code? and (ii) How do human and LLM preferences align with verifiable code properties and developer code tastes? To this end, we propose CodeFavor, a framework for training pairwise code preference models from synthetic evolution data, including code commits and code critiques. To evaluate code preferences, we introduce CodePrefBench, a benchmark comprising 1364 rigorously curated code preference tasks to cover three verifiable properties-correctness, efficiency, and security-along with human preference. Our evaluation shows that CodeFavor holistically improves the accuracy of model-based code preferences by up to 28.8%. Meanwhile, CodeFavor models can match the performance of models with 6-9x more parameters while being 34x more cost-effective. We also rigorously validate the design choices in CodeFavor via a comprehensive set of controlled experiments. Furthermore, we discover the prohibitive costs and limitations of human-based code preference: despite spending 23.4 person-minutes on each task, 15.1-40.3% of tasks remain unsolved. Compared to model-based preference, human preference tends to be more accurate under the objective of code correctness, while being sub-optimal for non-functional objectives.

Identifying Knowledge Editing Types in Large Language Models

Knowledge editing has emerged as an efficient approach for updating the knowledge of large language models (LLMs), attracting increasing attention in recent research. However, there is a notable lack of effective measures to prevent the malicious misuse of this technology, which could lead to harmful edits in LLMs. These malicious modifications have the potential to cause LLMs to generate toxic content, misleading users into inappropriate actions. To address this issue, we introduce a novel task, \textbf{K}nowledge \textbf{E}diting \textbf{T}ype \textbf{I}dentification (KETI), aimed at identifying malicious edits in LLMs. As part of this task, we present KETIBench, a benchmark that includes five types of malicious updates and one type of benign update. Furthermore, we develop four classical classification models and three BERT-based models as baseline identifiers for both open-source and closed-source LLMs. Our experimental results, spanning 42 trials involving two models and three knowledge editing methods, demonstrate that all seven baseline identifiers achieve decent identification performance, highlighting the feasibility of identifying malicious edits in LLMs. Additional analyses reveal that the performance of the identifiers is independent of the efficacy of the knowledge editing methods and exhibits cross-domain generalization, enabling the identification of edits from unknown sources. All data and code are available in https://github.com/xpq-tech/KETI. Warning: This paper contains examples of toxic text.

An Advanced Microscopic Energy Consumption Model for Automated Vehicle:Development, Calibration, Verification

The automated vehicle (AV) equipped with the Adaptive Cruise Control (ACC) system is expected to reduce the fuel consumption for the intelligent transportation system. This paper presents the Advanced ACC-Micro (AA-Micro) model, a new energy consumption model based on micro trajectory data, calibrated and verified by empirical data. Utilizing a commercial AV equipped with the ACC system as the test platform, experiments were conducted at the Columbus 151 Speedway, capturing data from multiple ACC and Human-Driven (HV) test runs. The calibrated AA-Micro model integrates features from traditional energy consumption models and demonstrates superior goodness of fit, achieving an impressive 90% accuracy in predicting ACC system energy consumption without overfitting. A comprehensive statistical evaluation of the AA-Micro model's applicability and adaptability in predicting energy consumption and vehicle trajectories indicated strong model consistency and reliability for ACC vehicles, evidenced by minimal variance in RMSE values and uniform RSS distributions. Conversely, significant discrepancies were observed when applying the model to HV data, underscoring the necessity for specialized models to accurately predict energy consumption for HV and ACC systems, potentially due to their distinct energy consumption characteristics.

V2X-VLM: End-to-End V2X Cooperative Autonomous Driving Through Large Vision-Language Models

Advancements in autonomous driving have increasingly focused on end-to-end (E2E) systems that manage the full spectrum of driving tasks, from environmental perception to vehicle navigation and control. This paper introduces V2X-VLM, an innovative E2E vehicle-infrastructure cooperative autonomous driving (VICAD) framework with large vision-language models (VLMs). V2X-VLM is designed to enhance situational awareness, decision-making, and ultimate trajectory planning by integrating data from vehicle-mounted cameras, infrastructure sensors, and textual information. The strength of the comprehensive multimodel data fusion of the VLM enables precise and safe E2E trajectory planning in complex and dynamic driving scenarios. Validation on the DAIR-V2X dataset demonstrates that V2X-VLM outperforms existing state-of-the-art methods in cooperative autonomous driving.

VLM-MPC: Vision Language Foundation Model (VLM)-Guided Model Predictive Controller (MPC) for Autonomous Driving

Motivated by the emergent reasoning capabilities of Vision Language Models (VLMs) and its potential to improve the comprehensibility of autonomous driving systems, this paper introduces a closed-loop autonomous driving controller called VLM-MPC, which combines a VLM for high-level decision-making and a Model Predictive Controller (MPC) for low-level vehicle control. The proposed VLM-MPC system is structurally divided into two asynchronous components: an upper-level VLM and a lower-level MPC. The upper layer VLM generates driving parameters for lower-level control based on front camera images, ego vehicle state, traffic environment conditions, and reference memory. The lower-level MPC controls the vehicle in real-time using these parameters, considering engine lag and providing state feedback to the entire system. Experiments based on the nuScenes dataset validated the effectiveness of the proposed VLM-MPC system across various scenarios (e.g., night, rain, intersections). Results showed that the VLM-MPC system consistently outperformed baseline models in terms of safety and driving comfort. By comparing behaviors under different weather conditions and scenarios, we demonstrated the VLM's ability to understand the environment and make reasonable inferences.