Association for the Advancement of Artificial Intelligence: AAAI Publications
Not a member yet
26155 research outputs found
Sort by
Reimagining Human-AI Social Collaboration
No full textAI increasingly mediates human social life—from daily communication and romantic partnerships to even "resurrecting" deceased loved ones. As AI’s influence on shaping human relationships and generative AI’s (GenAI) social capabilities grows, opportunities and challenges arise. In this work, we outline three roles AI could play in human-AI social collaboration and three levels of AI capabilities needed to perform various social roles effectively. We also discuss the measurement challenges and ethical risks involved. We aim to spark dialogue on designing sociotechnical systems that prioritize human social flourishing by examining AI’s role in mediating, simulating, and reshaping human social interaction
Challenges in Human-Compatible AI for Well-being: Harnessing Potential of GenAI for AI-Powered Science
No full textAt the AAAI Spring Symposium 2025, we explored the challenges of integrating Human-Compatible AI and AI-Powered Science to enhance social and individual well-being. Our discussion was guided by two perspectives.
Individual Impact of AI on Well-being: This perspective examines how AI influences personal autonomy, mental health, and emotional fulfillment. It seeks to ensure that AI enhances individual agency, rather than undermining critical thinking and independence.
Social Impact of AI on Well-being: This perspective focuses on the broader societal implications of AI, including fairness, misinformation, and economic impact. It emphasizes AI's role in fostering inclusive social structures while mitigating risks, such as bias and automation-driven unemployment.
This paper provides an overview of the motivations driving our exploration, defines key concepts, outlines major research challenges, and proposes strategies for integrating Human-Compatible AI and AI-Powered Science in a manner that balances innovation with ethical responsibility
Probabilistic Foundations for Metacognition via Hybrid-AI
No full textMetacognition is the concept of reasoning about an agent's own internal processes, and it has recently received renewed attention with respect to artificial intelligence (AI) and, more specifically, machine learning systems. This paper reviews a hybrid-AI approach known as "error detecting and correcting rules" (EDCR) that allows for the learning of rules to correct perceptual (e.g., neural) models. Additionally, we introduce a probabilistic framework that adds rigor to prior empirical studies, and we use this framework to prove results on necessary and sufficient conditions for metacognitive improvement, as well as limits to the approach. A set of future research directions is also provided
Tree-Based Approaches for Interpretable Modeling in Healthcare
No full textSurvival analysis models time-to-event data in oncology, such as cancer relapse or death, to evaluate treatment effects. Machine learning (ML) advancements, particularly ensemble methods like random survival forests (RSF), enhance predictive accuracy but often lack interpretability, posing challenges for clinical trust and regulatory compliance. This work addresses these limitations by systematically reviewing health authority criteria for AI interpretability, assessing existing methods like SurvSHAP and SurvLIME, and developing an RSF extension to handle multiple clinical events with novel metrics for performance evaluation. Future efforts focus on integrating model-specific interpretability through TreeSHAP and SurvSHAP to provide robust, time-dependent explanations, enabling the alignment of predictive power with clinical transparency in oncology care
Beyond Single Emotion: Multi-label Approach to Conversational Emotion Recognition
No full textEmotion recognition in conversation (ERC) has been promoted with diverse approaches in the recent years. However, many studies have pointed out that emotion shift and confusing labels make it difficult for models to distinguish between different emotions. Existing ERC models suffer from these problems when the emotions are forced to be mapped into single label. In this paper, we utilize our strategies for extending single label to multi-labels. We then propose a multi-label classification framework for emotion recognition in conversation (ML-ERC). Specifically, we introduce weighted supervised contrastive learning tailored for multi-label, which can easily applied to previous ERC models. The empirical results on existing task with single label support the efficacy of our approach, which is more effective in the most challenging settings: emotion shift or confusing labels. We also evaluate ML-ERC with the multi-labels we produced to support our contrastive learning scheme
Uncertainty-Aware Self-Training for CTC-Based Automatic Speech Recognition
No full textUncertainty estimation has been widely applied for trustworthy automatic speech recognition (ASR) systems across training and inference stages. In the training stage, previous studies show that uncertainty can facilitate self-training by filtering out unlabeled data samples with high uncertainty. However, the current sequence-level uncertainty estimation method for connectionist temporal classification (CTC) based ASR models drops the output probability information and depends only on the textual distance of decoded predictions. In this study, we argue that this results in limited performance improvement and propose a novel output probability-based sequence-level uncertainty estimation method. We also categorize uncertainty as pseudo-label uncertainty and in-training uncertainty for the self-training process. Finally, we present uncertainty-aware self-training for CTC-based ASR models and experimentally show the effectiveness of the proposed method compared to the baselines
KAES: Multi-aspect Shared Knowledge Finding and Aligning for Cross-prompt Automated Scoring of Essay Traits
No full textCross-prompt automated essay scoring (AES) aims to train models using essays from different source prompts and test them on new target prompt essays. A core challenge of the task is to learn as much shared knowledge as possible between essays from different prompts in order to better represent new prompt essays. Previous studies primarily focus on learning this knowledge on a general, coarse-grained level, ignoring that the shared knowledge among prompts is highly detailed and contains a more comprehensive range of information that is not fully investigated. In this paper, we propose a novel multi-aspect knowledge finding and aligning optimization strategy to better acquire this detailed various shared knowledge. We also introduce LLM to extract explicit, interpretable knowledge from implicit, multi-aspect shared knowledge and use this knowledge to improve the representation and evaluation performance of new prompt essays. We conduct extensive experiments on public datasets. The results show that our approach outperforms current state-of-the-art models and is effective on cross-prompt AES
How Do Position Encodings Affect Length Generalization? Case Studies On In-Context Function Learning
No full textThe capability of In-Context Learning (ICL) is crucial for large language models to generalize across a wide range of tasks. By utilizing prompts, these models can accurately predict outcomes for previously unseen tasks without necessitating retraining. However, this generalization ability does not extend to the length of the inputs; the effectiveness of ICL likely diminishes with excessively long inputs, resulting in errors in the generated text. To investigate this issue, we propose a study using a dataset of In-Context functions to understand the operational mechanisms of Transformer models in ICL and length generalization. We generated data using regression and Boolean functions and employed meta-learning techniques to endow the model with ICL capabilities. Our experimental results indicate that position encodings can significantly mitigate length generalization issues, with the most effective encoding extending the maximum input length to over eight times that of the original training length. However, further analysis revealed that while position encoding enhances length generalization, it compromises the model's inherent capabilities, such as its ability to generalize across different data types. Overall, our research illustrates that position encodings have a pronounced positive effect on length generalization, though it necessitates a careful trade-off with data generalization performance
Explore What LLM Does Not Know in Complex Question Answering
No full textComplex question answering (QA) is a challenging task in artificial intelligence research which requires reasoning based on related knowledge. The retrieval-augmented generation (RAG) based on large language models (LLMs) have become one promising solution in QA. To facilitate RAG more effectively, the LLM needs to precisely evaluate knowledge required in QA. That is, first, the LLM needs to examine its knowledge boundary (what the LLM does not know) to retrieve external knowledge as supplement. Second, the LLM needs to evaluate the utility of the retrieved knowledge (whether it helps in reasoning) for robust RAG. To this end, in this paper, we propose a novel Question Answering with Knowledge Evaluation (KEQA) framework to promote the effectiveness and efficiency of RAG in QA. First, inspired by quizzes in classroom, we propose a quiz-based method to precisely examine the knowledge state of the uninterpretable LLM for QA. We ask indicative quizzes on each required knowledge, and inspect whether the LLM can consistently answer the quiz to examine its knowledge boundary. Second, we retrieve the unknown knowledge from external source, and evaluate its utility to pick the helpful ones for reasoning. We design a reasoning-based metric to evaluate utility, and construct a demonstration set in training data for reference to guide knowledge picking in inference. We conduct extensive experiments on four widely-used QA datasets, and the results demonstrate the effectiveness of the proposed method
Relation Also Knows: Rethinking the Recall and Editing of Factual Associations in Auto-Regressive Transformer Language Models
No full textThe storage and recall of factual associations in auto-regressive transformer language models (LMs) have drawn a great deal of attention, inspiring knowledge editing by directly modifying the located model weights. Most editing works achieve knowledge editing under the guidance of existing interpretations of knowledge recall that mainly focus on subject knowledge. However, these interpretations are seriously flawed, neglecting relation information and leading to the *over-generalizing* problem for editing. In this work, we discover a novel relation-focused perspective to interpret the knowledge recall of transformer LMs during inference and apply it on single knowledge editing to avoid over-generalizing. Experimental results on the dataset supplemented with a new R-Specificity criterion demonstrate that our editing approach significantly alleviates over-generalizing while remaining competitive on other criteria, breaking the domination of subject-focused editing for future research