Association for the Advancement of Artificial Intelligence: AAAI Publications
Not a member yet
26155 research outputs found
Sort by
Fast and Interpretable Mixed-Integer Linear Program Solving by Learning Model Reduction
No full textBy exploiting the correlation between the structure and the solution of Mixed-Integer Linear Programming (MILP), Machine Learning (ML) has become a promising method for solving large-scale MILP problems. Existing ML-based MILP solvers mainly focus on end-to-end solution learning, which suffers from the scalability issue due to the high dimensionality of the solution space. Instead of directly learning the optimal solution, this paper aims to learn a reduced and equivalent model of the original MILP as an intermediate step. The reduced model often corresponds to interpretable operations and is much simpler, enabling us to solve large-scale MILP problems much faster than existing commercial solvers. However, current approaches rely only on the optimal reduced model, overlooking the significant preference information of all reduced models. To address this issue, this paper proposes a preference-based model reduction learning method, which considers the relative performance (i.e., objective cost and constraint feasibility) of all reduced models on each MILP instance as preferences. We also introduce an attention mechanism to capture and represent preference information, which helps improve the performance of model reduction learning tasks. Moreover, we propose a SetCover based pruning method to control the number of reduced models (i.e., labels), thereby simplifying the learning process. Evaluation on real-world MILP problems shows that 1) compared to the state-of-the-art model reduction ML methods, our method obtains nearly 20% improvement on solution accuracy, and 2) compared to the commercial solver Gurobi, two to four orders of magnitude speedups are achieved
Better Understandings and Configurations in MaxSAT Stochastic Local Search Solvers via Anytime Performance Analysis
No full textThough numerous solvers have been proposed for the MaxSAT problem, and the benchmark environment such as MaxSAT Evaluations provides a platform for the comparison of the state-of-the-art solvers, existing assessments were usually evaluated based on the quality, e.g., fitness, of the best-found solutions obtained within a given running time budget. However, concerning solely the final obtained solutions regarding specific time budgets may restrict us from comprehending the behavior of the solvers along the convergence process. This paper demonstrates that Empirical Cumulative Distribution Functions can be used to compare MaxSAT stochastic local search solvers' anytime performance across multiple problem instances and various time budgets. The assessment reveals distinctions in solvers' performance and displays that the (dis)advantages of solvers adjust along different running times. This work also exhibits that the quantitative and high variance assessment of anytime performance can guide machines, i.e., automatic configurators, to search for better parameter settings. Our experimental results show that the hyperparameter optimization tool, i.e., SMAC, can achieve better parameter settings of solvers when using the anytime performance as the cost function, compared to using the metrics based on the fitness of the best-found solutions
Token Highlighter: Inspecting and Mitigating Jailbreak Prompts for Large Language Models
No full textLarge Language Models (LLMs) are increasingly being integrated into services such as ChatGPT to provide responses to user queries. To mitigate potential harm and prevent misuse, there have been concerted efforts to align the LLMs with human values and legal compliance by incorporating various techniques, such as Reinforcement Learning from Human Feedback (RLHF), into the training of the LLMs. However, recent research has exposed that even aligned LLMs are susceptible to adversarial manipulations known as Jailbreak Attacks. To address this challenge, this paper proposes a method called Token Highlighter to inspect and mitigate the potential jailbreak threats in the user query. Token Highlighter introduced a concept called Affirmation Loss to measure the LLM's willingness to answer the user query. It then uses the gradient of Affirmation Loss for each token in the user query to locate the jailbreak-critical tokens. Further, Token Highlighter exploits our proposed Soft Removal technique to mitigate the jailbreak effects of critical tokens via shrinking their token embeddings. Experimental results on two aligned LLMs (LLaMA-2 and Vicuna-V1.5) demonstrate that the proposed method can effectively defend against a variety of Jailbreak Attacks while maintaining competent performance on benign questions of the AlpacaEval benchmark. In addition, Token Highlighter is a cost-effective and interpretable defense because it only needs to query the protected LLM once to compute the Affirmation Loss and can highlight the critical tokens upon refusal
JailPO: A Novel Black-Box Jailbreak Framework via Preference Optimization Against Aligned LLMs
No full textLarge Language Models (LLMs) aligned with human feedback have recently garnered significant attention. However, it remains vulnerable to jailbreak attacks, where adversaries manipulate prompts to induce harmful outputs. Exploring jailbreak attacks enables us to investigate the vulnerabilities of LLMs and further guides us in enhancing their security. Unfortunately, existing techniques mainly rely on handcrafted templates or generated-based optimization, posing challenges in scalability, efficiency and universality. To address these issues, we present JailPO, a novel black-box jailbreak framework to examine LLM alignment. For scalability and universality, JailPO meticulously trains attack models to automatically generate covert jailbreak prompts. Furthermore, we introduce a preference optimization-based attack method to enhance the jailbreak effectiveness, thereby improving efficiency. To analyze model vulnerabilities, we provide three flexible jailbreak patterns. Extensive experiments demonstrate that JailPO not only automates the attack process while maintaining effectiveness but also exhibits superior performance in efficiency, universality, and robustness against defenses compared to baselines. Additionally, our analysis of the three JailPO patterns reveals that attacks based on complex templates exhibit higher attack strength, whereas covert question transformations elicit riskier responses and are more likely to bypass defense mechanisms
Exploring Intrinsic Alignments Within Text Corpus
No full textRecent years have witnessed rapid advancements in the safety alignments of large language models (LLMs). Methods such as supervised instruction fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) have thus emerged as vital components in constructing LLMs. While these methods achieve robust and fine-grained alignment to human values, their practical application is still hindered by high annotation costs and incomplete human alignments. Besides, the intrinsic human values within training corpora have not been fully exploited. To address these issues, we propose ISAAC (Intrinsically Supervised Alignments by Assessing Corpus), a primary and coarse-grained safety alignment strategy for LLMs. ISAAC only relies on a prior assumption about the text corpus, and does not require preferences in RLHF or human responses selection in SFT. Specifically, it assumes a long-tail distribution of text corpus and employs a specialized sampling strategy to automatically sample high-quality responses. Theoretically, we prove that this strategy can improve the safety of LLMs under our assumptions. Empirically, our evaluations on mainstream LLMs show that ISAAC achieves a safety score comparable to current SFT solutions. Moreover, we conduct experiments on ISAAC for some RLHF-based LLMs, where we find that ISAAC can even improve the safety of these models under specific safety domains. These findings demonstrate that ISAAC can provide preliminary alignment to LLMs, thereby reducing the construction costs of existing human-feedback-based methods
Sequential Preference Optimization: Multi-Dimensional Preference Alignment with Implicit Reward Modeling
No full textHuman preference alignment is critical in building powerful and reliable large language models (LLMs). However, current methods either ignore the multi-dimensionality of human preferences (e.g. helpfulness and harmlessness) or struggle with the complexity of managing multiple reward models. To address these issues, we propose Sequential Preference Optimization (SPO), a method that sequentially fine-tunes LLMs to align with multiple dimensions of human preferences. SPO avoids explicit reward modeling, directly optimizing the models to align with nuanced human preferences. We theoretically derive closed-form optimal SPO policy and loss function. Gradient analysis is conducted to show how SPO manages to fine-tune the LLMs while maintaining alignment on previously optimized dimensions. Empirical results on LLMs of different size and multiple evaluation datasets demonstrate that SPO successfully aligns LLMs across multiple dimensions of human preferences and significantly outperforms the baselines
Neural Continuous-Time Supermartingale Certificates
No full textWe introduce for the first time a neural-certificate framework for continuous-time stochastic dynamical systems. Autonomous learning systems in the physical world demand continuous-time reasoning, yet existing learnable certificates for probabilistic verification assume discretization of the time continuum. Inspired by the success of training neural Lyapunov certificates for deterministic continuous-time systems and neural supermartingale certificates for stochastic discrete-time systems, we propose a framework that bridges the gap between continuous-time and probabilistic neural certification for dynamical systems under complex requirements. Our method combines machine learning and symbolic reasoning to produce formally certified bounds on the probabilities that a nonlinear system satisfies specifications of reachability, avoidance, and persistence. We present both the theoretical justification and the algorithmic implementation of our framework and showcase its efficacy on popular benchmarks
SEAL: Systematic Error Analysis for Value ALignment
No full textReinforcement Learning from Human Feedback (RLHF) aligns language models (LMs) with human values by training reward models (RMs) on binary preferences and using these RMs to fine-tune the base models. Despite its importance, the internal mechanisms of RLHF remain poorly understood. This paper introduces new metrics to evaluate RM effectiveness, focusing on feature imprint, feature resistance, and feature robustness. We categorize alignment datasets into target features (desired values) and spoiler features (undesired concepts). By regressing RM scores against these features, we quantify the extent to which RMs reward them -- feature imprint. We define alignment resistance as the proportion of the preference dataset where RMs fail to match human preferences, and we assess alignment robustness by analyzing RM responses to slightly perturbed texts. Our experiments, utilizing open-source components like the Anthropic/hh-rlhf preference dataset and OpenAssistant RMs, reveal significant imprints of target features and a notable sensitivity to spoiler features. We observed a 26% resistance incidence in portions of the dataset where LM labelers disagreed with human preferences. We also find that misalignment stems from confusing entries in the alignment dataset. These findings underscore the importance of scrutinizing both RMs and alignment datasets for a deeper understanding of value alignment
Align-Pro: A Principled Approach to Prompt Optimization for LLM Alignment
No full textThe alignment of large language models (LLMs) with human values is critical as these models become increasingly integrated into various societal and decision-making processes. Traditional methods, such as reinforcement learning from human feedback (RLHF), achieve alignment by fine-tuning model parameters, but these approaches are often computationally expensive and impractical when models are frozen or inaccessible for parameter modification. In contrast, prompt optimization is a viable alternative to RLHF for LLM alignment. While the existing literature has shown empirical promise of prompt optimization, its theoretical underpinning remains under-explored. We address this gap by formulating prompt optimization as an optimization problem and try to provide theoretical insights into the optimality of such a framework. To analyze the performance of the prompt optimization, we study theoretical suboptimality bounds and provide insights in terms of how prompt optimization depends upon the given prompter and target model. We also provide empirical validation through experiments on various datasets, demonstrating that prompt optimization can effectively align LLMs, even when parameter fine-tuning is not feasible
Towards a Theory of AI Personhood
No full textI am a person and so are you. Philosophically we sometimes
grant personhood to non-human animals, and entities such
as sovereign states or corporations can legally be considered
persons. But when, if ever, should we ascribe personhood to
AI systems? In this paper, we outline necessary conditions
for AI personhood, focusing on agency, theory-of-mind, and
self-awareness. We discuss evidence from the machine learn-
ing literature regarding the extent to which contemporary AI
systems, such as language models, satisfy these conditions,
finding the evidence surprisingly inconclusive.
If AI systems can be considered persons, then typical fram-
ings of AI alignment may be incomplete. Whereas agency has
been discussed at length in the literature, other aspects of per-
sonhood have been relatively neglected. AI agents are often
assumed to pursue fixed goals, but AI persons may be self-
aware enough to reflect on their aims, values, and positions
in the world and thereby induce their goals to change. We
highlight open research directions to advance the understand-
ing of AI personhood and its relevance to alignment. Finally,
we reflect on the ethical considerations surrounding the treat-
ment of AI systems. If AI systems are persons, then seeking
control and alignment may be ethically untenable