Large language models (LLMs) have shown great potential in complex reasoning tasks, yet their performance is often hampered by the scarcity of high-quality, reasoning-focused training datasets. Addressing this challenge, we propose Key-Point-Driven Data Synthesis (KPDDS), a novel data synthesis framework that synthesizes question-answer pairs by leveraging key points and exemplar pairs from authentic data sources. KPDDS ensures the generation of novel questions with rigorous quality control and substantial scalability. As a result, we present KPMath, the most extensive synthetic dataset tailored for mathematical reasoning to date, comprising over one million question-answer pairs. Utilizing KPMath and augmenting it with additional reasoning-intensive corpora, we create the comprehensive KPMath-Plus dataset. Fine-tuning the Mistral-7B model on KPMath-Plus yields a zero-shot PASS@1 accuracy of 39.3% on the MATH test set, a performance that not only outpaces other finetuned 7B models but also exceeds that of certain 34B models. Our ablation studies further confirm the substantial enhancement in mathematical reasoning across various subtopics, marking a significant stride in LLMs' reasoning capabilities.
Autonomous lane-change, a key feature of advanced driver-assistance systems, can enhance traffic efficiency and reduce the incidence of accidents. However, safe driving of autonomous vehicles remains challenging in complex environments. How to perform safe and appropriate lane change is a popular topic of research in the field of autonomous driving. Currently, few papers consider the safety of reinforcement learning in autonomous lane-change scenarios. We introduce safe hybrid-action reinforcement learning into discretionary lane change for the first time and propose Parameterized Soft Actor-Critic with PID Lagrangian (PASAC-PIDLag) algorithm. Furthermore, we conduct a comparative analysis of the Parameterized Soft Actor-Critic (PASAC), which is an unsafe version of PASAC-PIDLag. Both algorithms are employed to train the lane-change strategy of autonomous vehicles to output discrete lane-change decision and longitudinal vehicle acceleration. Our simulation results indicate that at a traffic density of 15 vehicles per kilometer (15 veh/km), the PASAC-PIDLag algorithm exhibits superior safety with a collision rate of 0%, outperforming the PASAC algorithm, which has a collision rate of 1%. The outcomes of the generalization assessments reveal that at low traffic density levels, both the PASAC-PIDLag and PASAC algorithms are proficient in attaining a 0% collision rate. Under conditions of high traffic flow density, the PASAC-PIDLag algorithm surpasses PASAC in terms of both safety and optimality.
We introduces Crimson, a system that enhances the strategic reasoning capabilities of Large Language Models (LLMs) within the realm of cybersecurity. By correlating CVEs with MITRE ATT&CK techniques, Crimson advances threat anticipation and strategic defense efforts. Our approach includes defining and evaluating cybersecurity strategic tasks, alongside implementing a comprehensive human-in-the-loop data-synthetic workflow to develop the CVE-to-ATT&CK Mapping (CVEM) dataset. We further enhance LLMs' reasoning abilities through a novel Retrieval-Aware Training (RAT) process and its refined iteration, RAT-R. Our findings demonstrate that an LLM fine-tuned with our techniques, possessing 7 billion parameters, approaches the performance level of GPT-4, showing markedly lower rates of hallucination and errors, and surpassing other models in strategic reasoning tasks. Moreover, domain-specific fine-tuning of embedding models significantly improves performance within cybersecurity contexts, underscoring the efficacy of our methodology. By leveraging Crimson to convert raw vulnerability data into structured and actionable insights, we bolster proactive cybersecurity defenses.
The growing integration of large language models (LLMs) into social operations amplifies their impact on decisions in crucial areas such as economics, law, education, and healthcare, raising public concerns about these models' discrimination-related safety and reliability. However, prior discrimination measuring frameworks solely assess the average discriminatory behavior of LLMs, often proving inadequate due to the overlook of an additional discrimination-leading factor, i.e., the LLMs' prediction variation across diverse contexts. In this work, we present the Prejudice-Caprice Framework (PCF) that comprehensively measures discrimination in LLMs by considering both their consistently biased preference and preference variation across diverse contexts. Specifically, we mathematically dissect the aggregated contextualized discrimination risk of LLMs into prejudice risk, originating from LLMs' persistent prejudice, and caprice risk, stemming from their generation inconsistency. In addition, we utilize a data-mining approach to gather preference-detecting probes from sentence skeletons, devoid of attribute indications, to approximate LLMs' applied contexts. While initially intended for assessing discrimination in LLMs, our proposed PCF facilitates the comprehensive and flexible measurement of any inductive biases, including knowledge alongside prejudice, across various modality models. We apply our discrimination-measuring framework to 12 common LLMs, yielding intriguing findings: i) modern LLMs demonstrate significant pro-male stereotypes, ii) LLMs' exhibited discrimination correlates with several social and economic factors, iii) prejudice risk dominates the overall discrimination risk and follows a normal distribution, and iv) caprice risk contributes minimally to the overall risk but follows a fat-tailed distribution, suggesting that it is wild risk requiring enhanced surveillance.
Existing large language models struggle to support numerous low-resource languages, particularly the extremely low-resource ones where there is minimal training data available for effective parameter updating. We thus investigate whether LLMs can learn a new language on the fly solely through prompting. To study this question, we collect a research suite for Zhuang, a language supported by no LLMs currently. We introduce \textsc{DiPMT++}, a framework for adapting LLMs to unseen languages by in-context learning. Using a dictionary and only 5K parallel sentences, \textsc{DiPMT++} significantly enhances the performance of GPT-4 from 0 to 16 BLEU for Chinese-to-Zhuang translation and achieves 32 BLEU for Zhuang-to-Chinese translation. Furthermore, we demonstrate the practical utility of this framework in aiding humans to translate completely unseen languages, which could contribute to the preservation of linguistic diversity.
Quantitative reasoning is a critical skill to analyze data, yet the assessment of such ability remains limited. To address this gap, we introduce the Quantitative Reasoning with Data (QRData) benchmark, aiming to evaluate Large Language Models' capability in statistical and causal reasoning with real-world data. The benchmark comprises a carefully constructed dataset of 411 questions accompanied by data sheets from textbooks, online learning materials, and academic papers. To compare models' quantitative reasoning abilities on data and text, we enrich the benchmark with an auxiliary set of 290 text-only questions, namely QRText. We evaluate natural language reasoning, program-based reasoning, and agent reasoning methods including Chain-of-Thought, Program-of-Thoughts, ReAct, and code interpreter assistants on diverse models. The strongest model GPT-4 achieves an accuracy of 58%, which has a large room for improvement. Among open-source models, Deepseek-coder-instruct, a code LLM pretrained on 2T tokens, gets the highest accuracy of 37%. Analysis reveals that models encounter difficulties in data analysis and causal reasoning, and struggle in using causal knowledge and provided data simultaneously. Code and data are in https://github.com/xxxiaol/QRData.
Manually-designed network architectures for thermal infrared pedestrian tracking (TIR-PT) require substantial effort from human experts. Neural networks with ResNet backbones are popular for TIR-PT. However, TIR-PT is a tracking task and more challenging than classification and detection. This paper makes an early attempt to search an optimal network architecture for TIR-PT automatically, employing single-bottom and dual-bottom cells as basic search units and incorporating eight operation candidates within the search space. To expedite the search process, a random channel selection strategy is employed prior to assessing operation candidates. Classification, batch hard triplet, and center loss are jointly used to retrain the searched architecture. The outcome is a high-performance network architecture that is both parameter- and computation-efficient. Extensive experiments proved the effectiveness of the automated method.
Blockchained Federated Learning (FL) has been gaining traction for ensuring the integrity and traceability of FL processes. Blockchained FL involves participants training models locally with their data and subsequently publishing the models on the blockchain, forming a Directed Acyclic Graph (DAG)-like inheritance structure that represents the model relationship. However, this particular DAG-based structure presents challenges in updating models with sensitive data, due to the complexity and overhead involved. To address this, we propose Blockchained Federated Unlearning (BlockFUL), a generic framework that redesigns the blockchain structure using Chameleon Hash (CH) technology to mitigate the complexity of model updating, thereby reducing the computational and consensus costs of unlearning tasks.Furthermore, BlockFUL supports various federated unlearning methods, ensuring the integrity and traceability of model updates, whether conducted in parallel or serial. We conduct a comprehensive study of two typical unlearning methods, gradient ascent and re-training, demonstrating the efficient unlearning workflow in these two categories with minimal CH and block update operations. Additionally, we compare the computation and communication costs of these methods.
This study focuses on a crucial task in the field of autonomous driving, autonomous lane change. Autonomous lane change plays a pivotal role in improving traffic flow, alleviating driver burden, and reducing the risk of traffic accidents. However, due to the complexity and uncertainty of lane-change scenarios, the functionality of autonomous lane change still faces challenges. In this research, we conduct autonomous lane-change simulations using both Deep Reinforcement Learning (DRL) and Model Predictive Control (MPC). Specifically, we propose the Parameterized Soft Actor-Critic (PASAC) algorithm to train a DRL-based lane-change strategy to output both discrete lane-change decision and continuous longitudinal vehicle acceleration. We also use MPC for lane selection based on predictive car-following costs for different lanes. For the first time, we compare the performance of DRL and MPC in the context of lane-change decision. Simulation results indicate that, under the same reward/cost functions and traffic flow, both MPC and PASAC achieve a collision rate of 0\%. PASAC demonstrates comparable performance to MPC in terms of episodic rewards/costs and average vehicle speeds.