Singapore

We consider the problem of risk-sensitive control in a reinforcement learning (RL) framework. In particular, we aim
to find a risk-optimal policy by maximizing the distortion
riskmetric (DRM) of the discounted reward in a finite-horizon Markov decision process (MDP). DRMs are a rich class
of risk measures that include several well-known risk measures as special cases. We derive a policy Hessian theorem
for the DRM objective using the likelihood ratio method. Using this result, we propose a natural DRM Hessian estimator from sample trajectories of the underlying MDP. Next,
we present a cubic-regularized policy Newton algorithm for
solving this problem in an on-policy RL setting using estimates of the DRM gradient and Hessian. Our proposed algorithm is shown to converge to an ϵ-second-order stationary
point (ϵ-SOSP) of the DRM objective, and this guarantee ensures the escaping of saddle points. The sample complexity
of our algorithms to find an ϵ-SOSP is O(ϵ−3.5). Our experiments validate the theoretical findings. To the best of our
knowledge, our is the first work to present convergence to an
ϵ-SOSP of a risk-sensitive objective, while existing works in
the literature have either shown convergence to a first-order
stationary point of a risk-sensitive objective, or a SOSP of a
risk-neutral one.

AAAI 2026

Policy Newton Methods for Distortion Riskmetrics

distortion riskmetrics

policy newton

risk-sensitive rl

The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. 

AAAI 2021

poster

We are pleased to announce the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26), which will be held in Singapore EXPO from January 20 to January 27, 2026.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-26 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.<br><br>

To access this event page, you need to log in with the **email address you registered with**. <br>Access credentials will be sent to your email from Underline -  subject line "Welcome to AAAI 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://aaai.org/conference/aaai/aaai-26/registration/) first.

**Online Registration Form**: https://aaai.getregistered.net/conference-2026 

Registration Required

We are pleased to announce the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26), which will be held in Singapore EXPO from January 20 to January 27, 2026.

Recently LLMs have faced increasing demands to selectively
remove specific information through Machine Unlearning. While evaluating unlearning effectiveness is crucial, exist-
ing benchmarks suffer from fundamental limitations in audit
dataset generation from unstructured corpora. We identify two
critical challenges: ensuring audit adequacy and handling knowledge redundancy between forget and retain datasets. Current approaches rely on ad-hoc question generation from
unstructured text, leading to unpredictable coverage gaps and
evaluation blind spots. Knowledge redundancy between for-get and retain corpora further obscures evaluation, making it difficult to distinguish genuine unlearning failures from
legitimately retained knowledge. To bring clarity to this chal-lenge, we propose LUCID, an automated framework that
leverages knowledge graphs to achieve comprehensive audit
dataset generation with fine-grained coverage and systematic redundancy elimination. By converting unstructured corpora into structured knowledge representations, it transforms the
ad-hoc audit dataset generation process into a transparent and
automated generation pipeline that ensures both adequacy
and non-redundancy. Applying LUCID to the MUSE bench-mark, we generated over 69,000 and 111,000 audit cases for
News and Books datasets respectively, identifying thousands of previously undetected knowledge memorization instances.
Our analysis reveals that knowledge redundancy significantly
skews metrics, artificially inflating ROUGE from 19.7% to
26.1% and Entailment Scores from 32.4% to 35.2%, highlight-
ing the necessity of deduplication for accurate assessment.

From Chaos to Clarity: A Knowledge Graph-Driven Audit Dataset Generation Framework for LLM Unlearning

Long-context processing remains a significant challenge for large language models (LLMs). Retrieval-augmented generation (RAG) has recently emerged as a promising approach, enabling LLMs to selectively access relevant information from extended contexts to improve efficiency. However, existing RAG approaches often lag behind other efficient long-context processing methods primarily due to inherent limitations on inaccurate retrieval and fragmented contexts. To address these limitations, we propose \textbf{RetroLM}, a novel RAG framework designed for effective long-context processing. Unlike traditional approaches, RetroLM introduces \textbf{KV-level retrieval augmentation}, which partitions the LLM's KV cache into contiguous pages and performs encoding and decoding operations based on the retrieved KV pages. Built upon this framework, we further develop a \textbf{specialized retriever} for precise retrieval of critical pages and conduct \textbf{unsupervised post-training} to optimize the model’s ability to leverage retrieved information. Compared with traditional RAG, the new approach enhances robustness to retrieval inaccuracy, facilitates effective utilization of fragmented contexts, and saves the cost from repeated context-encoding operations. We conduct extensive evaluations across several popular benchmarks, including LongBench, InfiniteBench, and RULER. RetroLM consistently outperforms existing long-LLMs and RAG-based methods, especially in tasks requiring deep reasoning or extreme context lengths. Our code and models will be released publicly to support future research in this area.

RetroLM: Retrieval-Augmented KVs for Long-Context Processing

Humans constantly generate a diverse range of tasks guided by internal motivations. While generative agents powered by large language models (LLMs) aim to simulate this complex behavior, it remains uncertain whether they operate on similar cognitive principles. To address this, we conducted a task-generation experiment comparing human responses with those of an LLM agent (GPT-4o). We find that human task generation is consistently influenced by psychological drivers, including personal values (e.g., Openness to Change) and cognitive style. Even when these psychological drivers are explicitly provided to the LLM, it fails to reflect the corresponding behavioral patterns. They produce tasks that are markedly less social, less physical, and thematically biased toward abstraction. Interestingly, while the LLM's tasks were perceived as more fun and novel, this highlights a disconnect between its linguistic proficiency and its capacity to generate human-like, embodied goals. We conclude that there is a core gap between the value-driven, embodied nature of human cognition and the statistical patterns of LLMs, highlighting the necessity of incorporating intrinsic motivation and physical grounding into the design of more human-aligned agents.

Mind the Gap: The Divergence Between Human and LLM-Generated Tasks

Vision-language object tracking overcomes the limitations of relying solely on visual features by leveraging language descriptions of objects to provide cross-modal semantic information, thereby enhancing model robustness in complex scenarios. However, most existing high-performance vision-language trackers are trained jointly on pure visual data and vision-language multimodal data. Due to the relative sparsity of language annotations in the data, the trackers tend to prioritize the localization role of visual features, diminishing the model's attention to language information. To mitigate this issue, we propose a novel vision-language tracker: Aware Distillation for Robust Vision-Language Tracking under Linguistic Sparsity (ADTrack). We introduce a knowledge distillation framework employing a knowledge-rich teacher model and a lightweight student model to establish modality correlations between vision and language, enabling efficient modeling between visual information and language descriptions. Specifically, our lightweight student module simultaneously distills language encoding capabilities from large language models through teacher-guided learning on input language, while performing target-aware perception on template images using language descriptions to generate more effective template features for subsequent visual extraction. Furthermore, to ensure perceptual robustness in linguistically sparse scenarios, we simulate language-deficient conditions during training and employ contrastive learning to enhance model adaptability. Extensive experiments demonstrate that ADTrack reduces parameters by over 50% while achieving state-of-the-art (SOTA) performance and speed on vision-language tracking benchmarks, including LaSOT, LaSOText, TNL2K, OTB-Lang and MGIT.

Aware Distillation for Robust Vision-Language Tracking Under Linguistic Sparsity

Mining time-frequency features is critical for time series forecasting. Existing research has predominantly focused on modeling low-frequency patterns, where most time series energy is concentrated. The overlooking of mid to high frequency continues to limit further performance gains in deep learning models. We propose FreqCycle, a novel framework integrating: (i) a Filter-Enhanced Cycle Forecasting (FECF) module to extract low-frequency features by explicitly learning shared periodic patterns in the time domain, and (ii) a Segmented Frequency-domain Pattern Learning (SFPL) module to enhance mid to high frequency energy proportion via learnable filters and adaptive weighting. Furthermore, time series data often exhibit coupled multi-periodicity, such as intertwined weekly and daily cycles. To address coupled multi-periodicity as well as long lookback window challenges, we extend FreqCycle hierarchically into MFreqCycle, which decouples nested periodic features through cross-scale interactions. Extensive experiments on seven diverse domain benchmarks demonstrate that FreqCycle achieves state-of-the-art accuracy while maintaining faster inference speeds, striking an optimal balance between performance and efficiency.

FreqCycle: A Multi-Scale Time-Frequency Analysis Method for Time Series Forecasting

Multi-view multi-label classification aims to utilize the rich information contained in multiple views for accurate classification. However, in real-world applications, its performance is often severely constrained by the concurrent missingness of both views and labels. To address this problem, this paper first targets the drawback of representation degradation in traditional feature disentanglement methods caused by strong consistency constraints and proposes a soft consistency constraint. This constraint not only effectively aligns the shared information and maximally avoids the compression of information beneficial to the classification task, but it also enhances the aggregation effect of high-quality representations on other representations. Furthermore, to address the coarse-grained problem of traditional fusion strategies, we designed a quality assessment network that achieves instance-level dynamic weighted fusion in a data-driven manner. Extensive experiments on multiple benchmark datasets demonstrate that our method achieves state-of-the-art performance in both incomplete and complete data scenarios, showcasing its robustness and generality.

Quality-aware and Soft Consistency Driven Representation Fusion for Incomplete Multi-view Multi-label Classification

In large-scale sensor networks, Multivariate Time Series Classification (MTSC) is a pivotal task for identifying events dependent on longitudinal data at the edge. However, existing methods focus on neither the inherent ability of convolutional networks to perceive subsequence features, nor the prolonged processing pipeline and the model deployment overhead brought by the highly parameterized models. To resolve these difficulties, we present EdgeMTSC, a lightweight large-kernel ConvNet for MTSC, which naturally extracts and learns features of diverse subsequences. Specifically, a novel module named Inter-channel Message Passing-driven Kernel Block (IMP-KB) is proposed, which maintains a learnable correlation matrix to propagate and merge inter-channel messages, and fuses miscellaneous patterns learned by parallel conv kernels of different sizes. EdgeMTSC sequences two modules of different receptive fields to aggregate local features using small kernels and study long-term representation provisioned by large kernels, respectively.
For inference parameter reduction and accelerating inference without performance loss, the conv blocks in IMP-KBs follows are structural re-reparameterizable. The performance of our model (76.2\%) is benchmarked on 26 UEA MTSC datasets and is superior to the SOTA model (MPTSNet, 75\%). At the same time, EdgeMTSC uses the least parameters and achieves the minimum inference time, applicable on any machine (8 devices ranging from large-scale distributed AI computing servers to resource-constrained edge devices) and in any application scenario.

EdgeMTSC: A Lightweight Large-Kernel ConvNet for Multivariate Time Series Classification

Evaluating a graphic design involves assessing it from multiple facets like alignment, composition, aesthetics and color choices. Holistic evaluation would involve aggregating feedback from individual expert reviewers. Towards this, we propose an Agentic Design Review System (Agentic-DRS), where multiple agents collaboratively analyze a design, orchestrated by a meta-agent. A novel in-context exemplar selection approach based on graph matching and a unique prompt expansion method plays central role towards making each agent design aware. In order to evaluate this framework, we propose DRS-BENCH. Thorough experimental evaluation against state-of-the-art baselines adapted to the problem setup, backed by critical ablations, demonstrates efficacy of Agentic-DRS in evaluating designs and generating actionable feedback.

Agentic Design Review System

Cytological images originate from exfoliated cells, collected via liquid-based slides and digitized into whole slide images (WSIs). Unlike histological WSIs that exhibit continuous and well-structured tissue, cytological WSIs are sparse in spatial distribution and unstructured in cellular relationships. Typically, the nucleus serves as the primary diagnostic feature, while surrounding cytoplasmic information plays a supportive role. These unique characteristics limit the development of effective foundation models and hinder the transferability of histology-based models for cytopathology. To address this, we propose **Cyto-SSL**, the first self-supervised pretraining framework for cytological images. It introduces **Nuclei-Centered Perturbation**, which highlights individual nuclei by perturbing non-nuclear regions. We also design an SR-Transformer module, which complements this by using sparse attention to concentrate on diagnostically relevant scattered cells, while iRPE helps model to capture local spatial relationships and avoids unnecessary attention to irrelevant global structures. Experimental results show that **Cyto-SSL** enhances performance across diverse cytological datasets and Multiple Instance Learning (MIL) methods. On a WSI-level dataset, it achieved 95.67\% accuracy and outperformed ImageNet-pretrained ResNet-50 by 11.33\%, demonstrating superior feature representation for cytological analysis. Additionally, **Cyto-SSL** modules are plug-and-play, easily integrated into other pretraining frameworks, yielding a 2.6\% accuracy gain across different SSL methods.

Cyto-SSL: A Self-Supervised Pretraining Framework for Cytology Foundation Model

Effective customer support requires not only accurate problem-solving but also structured and empathetic communication aligned with professional standards. However, existing dialogue datasets often lack strategic guidance, and real-world service data is difficult to access and annotate. To address this, we introduce the task of Customer Support Conversation (CSC), aimed at training customer service supporters to respond using well-defined support strategies. We propose a structured CSC framework grounded in COPC guidelines, defining five conversational stages and twelve strategies to guide high-quality interactions. Based on this, we construct CSConv, an evaluation dataset of 1,855 real-world customer–agent conversations rewritten using LLMs to reflect deliberate strategy use, and annotated accordingly. Additionally, we develop a role-playing approach that simulates strategy-rich conversations using LLM-powered roles aligned with the CSC framework, resulting in the training dataset RoleCS. Experiments show that fine-tuning strong LLMs on RoleCS significantly improves their ability to generate high-quality, strategy-aligned responses on CSConv. Human evaluations further confirm gains in problem resolution. All code and data will be publicly available at https://github.com/aliyun/qwen-dianjin.

Content not yet available

Next from AAAI 2026

From Chaos to Clarity: A Knowledge Graph-Driven Audit Dataset Generation Framework for LLM Unlearning

Similar lecture

Sample Complexity of Policy Gradient Finding Second-Order Stationary Points