Singapore

Self-play fine-tuning has emerged as a promising approach to improve Large Language Models (LLMs) without additional human annotations. However, existing methods struggle with complex generation tasks requiring long context understanding, where models produce partially correct outputs interleaved with errors. Traditional approaches train on entire sequences uniformly, failing to distinguish between well-predicted and erroneous regions, leading to diluted learning signals and slow convergence. We propose DRIFT (Difference-aware Reinforcement through Iterative Fine-Tuning), a novel self-play framework that selectively trains on prediction differences. DRIFT introduces two key innovations: (1) Difference-Aware Masking (DAM) that identifies and masks common subsequences between model outputs and ground truth, focusing training exclusively on error regions; (2) Occurrence-Aware Loss (OAL) that provides position-invariant vocabulary supervision, complementing the position-sensitive adversarial loss. This dual mechanism enables models to correct both positional and lexical errors effectively. Theoretically, we prove that DRIFT converges when masked distributions align. Empirically, we evaluate DRIFT on diverse summarization benchmarks using Qwen2.5-3B and LLaMA-3.1-8B models. Results show that DRIFT significantly outperforms both supervised fine-tuning (SFT) and self-play fine-tuning (SPIN), achieving up to 16\% improvement on SAMSum dialogue summarization tasks while maintaining general capabilities. Notably, DRIFT breaks the performance ceiling of continued SFT and demonstrates superior efficiency compared to holistic self-play methods, validating that targeted optimization on prediction differences is crucial for structured text generation tasks.

AAAI 2026

DRIFT: Difference-Aware Reinforcement Through Iterative Fine-Tuning for Language Model

large language model

self-play

summarization

generation

reinforcement learning

### Welcome to ACL 2023, the 61st Annual Meeting of the Association for Computational Linguistics! 
 The conference will be held in Toronto, Canada, July 9-14, 2023. 
Following the succession of the recent conferences in our field, ACL 2023 will adopt a hybrid format.
While the impact of Covid has considerably diminished in terms of traveling, obtaining visas to Canada
entails a very long process. Moreover, the global economic conditions pose challenges for many individuals to travel to conferences. Recognizing these circumstances, we know many participants may not be
able to attend the conference in person. Therefore, we are committed to providing a great virtual platform
so everyone has the opportunity to interact with other participants and enjoy the conference. Based on the
current registered participants, approxiately 30% have chosen to attend the conference virtually. Whether
you join us in person or virtually, we sincerely hope everyone has a remarkable conference experience. 
This General Chair’s message is where I express my gratitude to the many individuals who have made
enormous contributions to the conference over the past year.

Read [**ACL 2023 General Chair's message**](https://docs.google.com/document/d/1WobYM7norbG4dI48s75HfJoD89qgX5a_F-6U8AteLSA/edit?usp=sharing/) in full.

##### **[Conference Handbook](https://2023.aclweb.org/downloads/acl2023-handbook.pdf)**

ACL 2023

The Association for Computational Linguistics (ACL) is the premier international scientific and professional society for people working on computational problems involving human language, a field often referred to as either computational linguistics or natural language processing.

technical paper

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.

To access this event page, you need to log in with the **email address you registered with**. 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.

Establishing convergent semantics for weighted argumentation graph is a fundamental and long-standing issue. Particularly, it is challenging to develop convergent semantics for weighted bipolar argumentation graph (wBAG), which include both support and attack relations on weighted arguments. Existing semantics in the literature are not general enough in the sense that they only apply to acyclic graphs or special cyclic cases. In this paper, we provide a general and elegant solution to this issue by adopting the so-called bilateral gradual semantics, so that the strength of arguments can be defined as the limits of iterative functions that always converge for any wBAG including cyclic ones. A preliminary experimental analysis shows that our semantics appear quite efficient in calculating argument strength. Overall, this paper offers a solid and promising foundation for weighted bipolar argumentation in theoretical and practical aspects.

Convergent Semantics for Weighted Bipolar Argumentation

Medical image segmentation plays a critical role in clinical diagnosis, lesion quantification, and preoperative planning. However, conventional Mamba-based architectures, which rely on fixed-directional sequence modeling and flattening of images into one-dimensional sequences, fail to effectively capture hierarchical anatomical features and spatial dependencies, limiting their representation capacity for complex medical structures. To address these challenges, we propose EccoMamba (Enhanced Cross-hierarchical Continuity Orthogonal Mamba), a U-shaped encoder-decoder framework tailored for medical image segmentation. In the encoder’s downsampling path, we introduce the hierarchical aggregation enhancement (HAE) module, which integrates multi-scale convolutions with hierarchical attention. The attention branch further incorporates cross-channel interactions, enabling the model to amplify semantically relevant features and suppress background noise selectively. For skip connections, we design the structural continuity orthogonal (SCO) module to maintain spatial continuity by modeling cross-dimensional dependencies via orthogonal axial shifts, mitigating directional bias and enhancing anatomical consistency. Extensive experiments on four benchmark datasets, ISIC 2018, ISIC 2017, Synapse, and ACDC, demonstrate that EccoMamba consistently outperforms state-of-the-art methods in both segmentation accuracy and structural fidelity. The code is available at https://anonymous.4open.science/r/EccoMamba-4117/.

EccoMamba: Enhanced Cross-hierarchical Continuity Orthogonal Mamba for Medical Image Segmentation

Beyond scratch coding, exploiting large-scale code repositories (e.g., GitHub) for practical tasks is vital in real-world software development, yet current benchmarks rarely evaluate code agents in such authentic, workflow-driven scenarios. 
To bridge this gap, we introduce GitTaskBench, a benchmark designed to systematically assess this capability via 54 realistic tasks across 7 modalities and 7 domains. 
Each task pairs a relevant repository with an automated, human-curated evaluation harness specifying practical success criteria. Beyond measuring execution and task success, we also propose the alpha-value metric to quantify the economic benefit of agent performance, which integrates task success rates, token cost, and average developer salaries. Experiments across three state-of-the-art agent frameworks with multiple advanced LLMs show that leveraging code repositories for complex task solving remains challenging: even the best-performing system, OpenHands+Claude 3.7, solves only 48.15% of tasks (recent progress has pushed the frontier further, with RepoMaster+Claude 3.5 achieving a new record of 62.96%). Error analysis attributes over half of failures to seemingly mundane yet critical steps like environment setup and dependency resolution, highlighting the need for more robust workflow management and increased timeout preparedness.
By releasing GitTaskBench, we aim to drive progress and attention toward repository-aware code reasoning, execution, and deployment---moving agents closer to solving complex, end-to-end real-world tasks. The code & datasets are public in https://github.com/QuantaAlpha/GitTaskBench. See the extended version at https://arxiv.org/abs/2508.18993.

GitTaskBench: A Benchmark for Code Agents Solving Real-World Tasks Through Code Repository Leveraging

The core problem in multi-view anomaly detection is to represent local neighborhoods of normal instances consistently across all views. Recent approaches consider a representation of local neighborhood in each view independently, and then capture the consistent neighbors across all views via a learning process. They suffer from two key issues. First, there is no guarantee that they can capture consistent neighbors well, especially when the same neighbors are in regions of varied densities in different views, resulting in inferior detection accuracy. Second, the learning process has a high computational cost of $\mathcal{O}(N^2)$, rendering them inapplicable for large datasets. To address these issues, we propose a novel method termed \textbf{S}pherical \textbf{C}onsistent \textbf{N}eighborhoods \textbf{E}nsemble (SCoNE). It has two unique features: (a) the consistent neighborhoods are represented with multi-view instances directly, requiring no intermediate representations as used in existing approaches; and (b) the neighborhoods have data-dependent properties, which lead to large neighborhoods in sparse regions and small neighborhoods in dense regions. The data-dependent properties enable local neighborhoods in different views to be represented well as consistent neighborhoods, without learning. This leads to $\mathcal{O}(N)$ time complexity. Empirical evaluations show that SCoNE has superior detection accuracy and runs orders-of-magnitude faster in large datasets than existing approaches.

SCoNE: Spherical Consistent Neighborhoods Ensemble for Effective and Efficient Multi-View Anomaly Detection

We study strategic candidate nomination by parties in elections decided by Plurality voting. Each party selects a nominee before the election, and the winner is chosen from the nominated candidates based on the voters' preferences. We introduce a new restriction on these preferences, which we call party-aligned single-peakedness: all voters agree on a common ordering of the parties along an ideological axis, but may differ in their perceptions of the positions of individual candidates within each party. The preferences of each voter are single-peaked with respect to their own axis over the candidates, which is consistent with the global ordering of the parties.
We present a polynomial-time algorithm for recognizing whether a preference profile satisfies party-aligned single-peakedness. 
In this domain, we give polynomial-time algorithms for deciding whether a given party can become the winner under some (or all) nominations, and whether this can occur in some pure Nash equilibrium. Moreover, we prove a tight result about the guaranteed existence of pure strategy Nash equilibria for elections with up to three parties for single-peaked and party-aligned single-peaked preference profiles.

Computing Equilibrium Nominations in Presidential Elections

Test-time adaptation (TTA) enables online model adaptation using only unlabeled test data, aiming to bridge the gap between source and target distributions. However, in multimodal scenarios, varying degrees of distribution shift across different modalities give rise to a complex coupling effect of unimodal shallow feature shift and cross-modal high-level semantic misalignment, posing a major obstacle to extending existing TTA methods to the multimodal field. To address this challenge, we propose a novel multimodal test-time adaptation (MMTTA) framework, termed as Bridging Modalities via Progressive Re-alignment (BriMPR). BriMPR, consisting of two progressively enhanced modules, tackles the coupling effect with a divide-and-conquer strategy. Specifically, we first decompose MMTTA into multiple unimodal feature alignment sub-problems. By leveraging the strong function approximation ability of prompt tuning, we calibrate the unimodal global feature distributions to their respective source distributions, so as to achieve the initial semantic re-alignment across modalities. Subsequently, we assign the credible pseudo-labels to combinations of masked and complete modalities, and introduce inter-modal instance-wise contrastive learning to further enhance the information interaction among modalities and refine the alignment. Extensive experiments on MMTTA tasks, including both corruption-based and real-world domain shift benchmarks, demonstrate the superiority of our method.

Bridging Modalities via Progressive Re-alignment for Multimodal Test-Time Adaptation

Existing battery State of Health (SOH) prediction approaches often struggle to provide both accurate predictions and reliable uncertainty estimates. This paper presents a novel Multi-Task Learning (MTL) framework that jointly tackles SOH prediction and provides a proxy metric for uncertainty through a unified architecture. The framework combines a Physics-Informed Neural Network (PINN) for SOH prediction with a deep autoencoding Gaussian mixture model for uncertainty modeling. Particularly, the energy score from the Gaussian mixture model serves as a proxy metric for uncertainty, where a higher score indicates potential prediction unreliability. Moreover, to enhance task-specific learning, we employ a multi-head attention mechanism that adaptively captures distinct feature relationships. 
Our experiments show improvements in prediction performance compared to the state-of-the-art baseline.
A comprehensive evaluation on six XJTU battery benchmark datasets demonstrates that our framework achieves a prediction accuracy of 99.50% (MAPE: 0.0050) while providing reliable uncertainty quantification through the proxy metric.

Physics-Informed Multi-Task Learning for Battery State of Health Prediction with Uncertainty Quantification

High-resolution Earth Observation technologies present unprecedented opportunities for geospatial analysis, yet traditional 2D aerial-view semantic segmentation remains limited by its inability to model spatial relationships and handle object occlusions. While 3D Aerial-view Segmentation (3DAS) has emerged to address these limitations, existing methods predominantly rely on 2D discriminative models pre-trained on natural scenes. These models struggle to accurately recognize aerial-view imagery, resulting in suboptimal performance due to significant domain discrepancies. This paper introduces ID-Splat, a novel object-centric framework that directly leverages multi-view object identities without discriminative information to enhance 3D semantic understanding. ID-Splat implements a two-stage process: first, Mask-object Tracking combines SAM and Point Tracking to establish robust and consistent object identities across multi-view aerial images; second, Object Integration & Propagation assigns these identities to 3D Gaussian Splatting (3DGS) points, enabling complete 3D segmentation through semantic propagation. Experimental results on the 3D-AS dataset demonstrate that ID-Splat significantly outperforms existing methods, particularly under sparse supervision conditions. ID-Splat also achieves state-of-the-art performance while reducing the need for extensive labeled data by effectively leveraging the inherent 3D structure.

ID-Splat: Propagating Object Identities for Segmenting 3D Aerial-view Scenes

Continual Learning with Pre-trained Models holds great promise for efficient adaptation across sequential tasks. However, most existing approaches freeze PTMs and rely on auxiliary modules like prompts or adapters, limiting model plasticity and leading to suboptimal generalization when facing significant distribution shifts. While full fine-tuning can improve adaptability, it risks disrupting crucial pre-trained knowledge. In this paper, we propose Mutual Information-guided Sparse Tuning (MIST), a plug-and-play method that selectively updates a small subset of PTM parameters, less than 5%, based on sensitivity to mutual information objectives. MIST enables effective task-specific adaptation while preserving generalization. To further reduce interference, we introduce strong sparsity regularization by randomly dropping gradients during tuning, resulting in fewer than 0.5% of parameters being updated per step. Applied before standard freeze-based methods, MIST consistently boosts performance across diverse continual learning benchmarks. Experiments show that integrating our method into multiple baselines yields significant performance gains. Our code is available at \url{https://github.com/zhwhu/MIST}.

Sparse Tuning Enhances Plasticity in PTM-based Continual Learning

Gaussian process regression (GPR) is a popular nonparametric Bayesian method that provides predictive uncertainty estimates and is widely used in safety-critical applications. While prior research has introduced various uncertainty bounds, most existing approaches 
require access to specific input features and rely on posterior mean and variance estimates or tuning hyperparameters. These limitations hinder robustness and fail to capture the model’s global behavior in expectation.
To address these limitations, we propose a chaining-based framework for estimating upper and lower bounds on the expected extreme values over unseen data, without requiring access to specific input locations. We provide kernel-specific refinements for commonly used kernels such as RBF and Matérn, in which our bounds are tighter than generic constructions. We further improve numerical tightness by avoiding analytical relaxations. 
In addition to global estimation, we also develop a novel method for local uncertainty quantification at specified inputs. This approach leverages chaining geometry through partition diameters, adapting to local structure without relying on posterior variance scaling. 
Our experimental results validate the theoretical findings and demonstrate that our method outperforms existing approaches on both synthetic and real-world datasets.

Downloads

Next from AAAI 2026

Convergent Semantics for Weighted Bipolar Argumentation

Similar lecture

Characterizing and Measuring Linguistic Dataset Drift