Singapore

Recent advancements in multimodal large language models for video understanding (videoLLMs) have enhanced their capacity to process complex spatiotemporal data. However, challenges such as factual inaccuracies, harmful content, biases, hallucinations, and privacy risks compromise their reliability. This study introduces Trust-videoLLMs, a first comprehensive benchmark evaluating 23 state-of-the-art videoLLMs (5 commercial, 18 open-source) across five critical dimensions: truthfulness, robustness, safety, fairness, and privacy. Comprising 30 tasks with adapted, synthetic, and annotated videos, the framework assesses spatiotemporal risks, temporal consistency and cross-modal impact. Results reveal significant limitations in dynamic scene comprehension, cross-modal perturbation resilience and real-world risk mitigation. While open-source models occasionally outperform, proprietary models generally exhibit superior credibility, though scaling does not consistently improve performance. These findings underscore the need for enhanced training datat diversity and robust multimodal alignment. Trust-videoLLMs provides a publicly available, extensible toolkit for standardized trustworthiness assessments, addressing the critical gap between accuracy-focused benchmarks and demands for robustness, safety, fairness, and privacy.
Code are available at: \emph{https://anonymous.4open.science/r/Trust-videoLLMs-0C37}

AAAI 2026

Benchmarking Trustworthiness in Multimodal LLMs for Video Understanding

large langauge models

trustworthiness

video

safety

evaluation

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.

technical paper

Visual Place Recognition (VPR) aims to determine the geographic location of a query image by retrieving its most visually similar counterpart from a geo-tagged reference database. Recently, the emergence of the powerful visual foundation model, DINOv2, trained in a self-supervised manner on massive datasets, has significantly improved VPR performance. This improvement stems from DINOv2’s exceptional feature generalization capabilities but is often accompanied by increased model complexity and computational overhead that impede deployment on resource-constrained devices. To address this challenge, we propose $D^{2}$-VPR, a $D$istillation- and $D$eformable-based framework that retains the strong feature extraction capabilities of visual foundation models while significantly reducing model parameters and achieving a more favorable performance-efficiency trade-off. Specifically, first, we employ a two-stage training strategy that integrates knowledge distillation and fine-tuning. Additionally, we introduce a Distillation Recovery Module (DRM) to better align the feature spaces between the teacher and student models, thereby minimizing knowledge transfer losses to the greatest extent possible. Second, we design a Top-Down-attention-based Deformable Aggregator (TDDA) that leverages global semantic features to dynamically and adaptively adjust the Regions of Interest (ROI) used for aggregation, thereby improving adaptability to irregular structures. Extensive experiments demonstrate that our method achieves competitive and even superior performance compared to state-of-the-art approaches. Meanwhile, it reduces the parameter count by approximately 64.2\% and FLOPs by about 62.6\% (compared to CricaVPR).

D²-VPR: A Parameter-efficient Visual-foundation-model-based Visual Place Recognition Method via Knowledge Distillation and Deformable Aggregation

This paper studies the problem of minimizing group-level inequity in facility location games on the real line, where agents are partitioned into groups and may act strategically. While most classical objectives focus on minimizing total or maximum individual cost, we consider a fairness-driven objective that minimizes the maximum group effect (Marsh and Schilling 1994), where each group's effect is defined as its total or maximum distance to the nearest facility, weighted by group-specific factors. We show that this formulation generalizes several prominent optimization objectives, including the classical utilitarian (social cost) and egalitarian (maximum cost) objectives Procaccia and Tennenholtz (2009); and two group-fair objectives, maximum total/average group cost, proposed by Zhou, Li, and Chan (2022). 
To minimize the maximum group effect objective, for single-facility case, we propose novel BALANCED mechanism and Major-Phantom mechanism, both of which are strategyproof and provide tight approximation guarantees when considering distinct maximum group effect objectives. Our mechanisms not only resolve the bound gap for group-fairness objectives studied by Zhou, Li, and Chan (2022) but also unify many classical truthful mechanisms with respect to classical optimization objectives. For the two-facility case, we revisit the classical endpoint mechanism in our more generalized scenario and show that the mechanism is powerful to provide tight bounds for two distinct maximum group effect objectives.

Minimizing Inequity in Facility Location Games

Feature caching has recently emerged as a promising method for diffusion model acceleration. It effectively alleviates the inefficiency problem caused by high computational requirements by caching similar features in the inference process of the diffusion model. In this paper, we analyze existing feature caching methods from the perspective of information utilization, and point out that relying solely on historical information will lead to constrained accuracy and speed performance. And we propose a novel paradigm that introduces future information via self-speculation based on the information similarity at the same time step across different iteration times. Based on this paradigm, we present \textit{SpecDiff}, a training-free multi-level feature caching strategy including a cached feature selection algorithm and a multi-level feature classification algorithm. (1) Feature selection algorithm based on self-speculative information. \textit{SpecDiff} determines a dynamic importance score for each token based on self-speculative information and historical information, and performs cached feature selection through the importance score. (2) Multi-level feature classification algorithm based on feature importance scores. \textit{SpecDiff} classifies tokens by leveraging the differences in feature importance scores and introduces a multi-level feature calculation strategy.
Extensive experiments show that \textit{SpecDiff} achieves average 2.80×, 2.74×, and 3.17× speedup with negligible quality loss in Stable Diffusion 3, 3.5, and FLUX compared to RFlow on NVIDIA A800-80GB GPU. By merging speculative and historical information, \textit{SpecDiff} overcomes the speedup-accuracy trade-off bottleneck, pushing the Pareto frontier of speedup and accuracy in the efficient diffusion model inference.

SpecDiff: Accelerating Diffusion Model Inference with Self-Speculation

One crucial factor behind the success of deep learning lies in the implicit bias induced by noise inherent in gradient-based training algorithms. Motivated by empirical observations that training with noisy labels improves model generalization, we delve into the underlying mechanisms behind stochastic gradient descent (SGD) with label noise. Focusing on a two-layer over-parameterized linear network, we analyze the learning dynamics of label noise SGD, unveiling a two-phase learning behavior. In Phase I, the magnitudes of model weights progressively diminish, and the model escapes the lazy regime; enters the rich regime. In Phase II, the alignment between model weights and the ground-truth interpolator increases, and the model eventually converges. Our analysis highlights the critical role of label noise in driving the transition from the lazy to the rich regime and minimally explains its empirical success. Extensive experiments, conducted under both synthetic and real-world setups, strongly support our theory.

On the Learning Dynamics of Two-layer Linear Networks with Label Noise SGD

Long-term Time Series Forecasting is crucial across numerous critical domains, yet its accuracy remains fundamentally constrained by the receptive field bottleneck in existing models. Mainstream Transformer- and Multi-layer Perceptron (MLP)-based methods mainly rely on finite look-back windows, limiting their ability to model long-term dependencies and hurting forecasting performance. Naively extending the look-back window proves ineffective, as it not only introduces prohibitive computational complexity, but also drowns vital long-term dependencies in historical noise. To address these challenges, we propose CometNet, a novel Contextual Motif-guided Long-term Time Series Forecasting framework. CometNet first introduces a Contextual Motif Extraction module that identifies recurrent, dominant contextual motifs from complex historical sequences, providing extensive temporal dependencies far exceeding limited look-back windows; Subsequently, a Motif-guided Forecasting module is proposed, which integrates the extracted dominant motifs into forecasting. By dynamically mapping the look-back window to its relevant motifs, CometNet effectively harnesses their contextual information to strengthen long-term forecasting capability. Extensive experimental results on eight real-world datasets have demonstrated that CometNet significantly outperforms current state-of-the-art (SOTA) methods, particularly on extended forecast horizons.

CometNet: Contextual Motif-guided Long-term Time Series Forecasting

Dataset Condensation (DC) distills knowledge from large datasets into smaller ones, accelerating training and reducing storage requirements. However, despite notable progress, prior methods have largely overlooked the potential of quantization for further reducing storage costs. In this paper, we take the first step to explore post-training quantization in dataset condensation, demonstrating its effectiveness in reducing storage size while maintaining representation quality without requiring expensive training cost. However, we find that at extremely low bit-widths (e.g., 2-bit), conventional quantization leads to substantial degradation in representation quality, negatively impacting the networks trained on these data.
To address this, we propose a novel \emph{patch-based post-training quantization} approach that ensures localized quantization with minimal loss of information. To reduce the overhead of quantization parameters, especially for small patch sizes, we employ quantization-aware clustering to identify similar patches and subsequently aggregate them for efficient quantization. Furthermore, we introduce a refinement module to align the distribution between original images and their dequantized counterparts, compensating for quantization errors.
Our method is a plug-and-play framework that can be applied to synthetic images generated by various DC methods. Extensive experiments across diverse benchmarks including CIFAR-10/100, Tiny ImageNet, and ImageNet subsets demonstrate that our method consistently outperforms prior works under the same storage constraints. Notably, our method nearly \textbf{doubles the test accuracy} of existing methods at extreme compression regimes (e.g., 26.0\% $\rightarrow$ 54.1\% for DM at IPC=1), while operating directly on 2-bit images without additional distillation.

Post Training Quantization for Efficient Dataset Condensation

Selective deletion of data from deep models, known as unlearning, has become crucial for enforcing the right to be forgotten, while also mitigating the negative impact of flawed training data. Retraining deep models is often impractical due to data access restrictions and computational overhead. Existing retraining-free methods are typically based on the Fisher Information Matrix (FIM), which quantifies the importance of model parameters with respect to forgetting classes, applying equal dampening to these parameters. This approach implicitly assumes a semantically uniform representation space, where all retained classes are equidistant from the forgetting classes. However, this assumption often fails in real-world cross-modal retrieval scenarios characterized by multi-label and non-orthogonal semantics.
To overcome this limitation, we propose Prior-Prototype guided Partitioned dampening (PPP), an effective strategy for selective forgetting in cross-modal retrieval. First, PPP defines prior prototypes, which are semantic centers derived from well-trained models, to identify neighbor classes semantically close to the forgetting set. 
Then, PPP uses Fisher information to identify parameters sensitive to forgetting and partitions them into buffer and core regions based on their relative importance to the neighbor and retained sets.
Finally, PPP applies a hierarchical dampening strategy, where core parameters receive stronger suppression guided by prototype-based semantic disparities.
Comprehensive evaluations on four large-scale benchmarks show that PPP performs competitively with retraining-based baselines, highlighting its effectiveness and generalizability in selective unlearning for cross-modal retrieval.

Unlearning in Cross-Modal Retrieval via Prior-Prototype Guided Partitioned Dampening

Neural Speech Codecs face a fundamental trade-off at low bitrates: preserving acoustic fidelity often compromises semantic richness. To address this, we introduce SACodec, a novel codec built upon an asymmetric dual-quantizer that employs our proposed $\textbf{S}$emantic $\textbf{A}$nchoring mechanism. This design strategically decouples the quantization of $\textbf{S}$emantic and $\textbf{A}$coustic details. The semantic anchoring is achieved via a lightweight projector that aligns acoustic features with a frozen, large-scale mHuBERT codebook, injecting linguistic priors while guaranteeing full codebook utilization. Sequentially, for acoustic details, a residual activation module with SimVQ enables a single-layer quantizer (acoustic path) to faithfully recover fine-grained information. At just 1.5\,kbps, SACodec establishes a new state of the art by excelling in both fidelity and semantics: subjective listening tests confirm that its reconstruction quality is perceptually highly comparable to ground-truth audio,
while its tokens demonstrate substantially improved semantic richness in downstream tasks. This work suggests that assigning specialized semantic quantizers to distinct information streams offers an effective path to reconcile the long-standing trade-off between fidelity, semantics, and modeling simplicity in low-bitrate speech tokenization.

SACodec: Asymmetric Quantization with Semantic Anchoring for Low-Bitrate High-Fidelity Neural Speech Codecs

Drug discovery is a very time-consuming and costly endeavour due to its huge design space and to the lengthy and failure-fraught process of bringing a product to market. Automating the generation of candidate molecules exhibiting some of the desired properties can help. Among the standard formats to encode molecules, SMILES is a widespread string representation. We propose a constraint programming model showcasing the grammar constraint to express the design space of organic molecules using the SMILES notation. We show how some common physicochemical properties --- such as molecular weight and lipophilicity --- and structural features can be expressed as constraints in the model. We also contribute a weighted counting algorithm for the grammar constraint, allowing us to use a belief propagation heuristic to guide the generation. Our experiments indicate that such a heuristic is key to driving the search towards desired molecules.

Constrained Molecule Generation Modelled Using the Grammar Constraint

Large Language Models (LLMs) have achieved significant advances in natural language processing, yet their potential for high-stake political decision-making remains largely unexplored. This paper addresses the gap by focusing on the application of LLMs to the United Nations (UN) decision-making process, where the stakes are particularly high and political decisions can have far-reaching consequences. We introduce a novel dataset comprising publicly available UN Security Council (UNSC) records from 1994 to 2024, including draft resolutions, voting records, and diplomatic speeches. Using this dataset, we propose the United Nations Benchmark (UNBench), the first comprehensive benchmark designed to evaluate LLMs across four interconnected political science tasks: co-penholder judgment, representative voting simulation, draft adoption prediction, and representative statement generation. These tasks span the three stages of the UN decision-making process—drafting, voting, and discussing—and aim to assess LLMs' ability to understand and simulate political dynamics. Our experimental analysis demonstrates the potential and challenges of applying LLMs in this domain, providing insights into their strengths and limitations in political science. To the best of our knowledge, this is the first benchmark to systematically evaluate LLMs in UN decision-making, contributing to the growing intersection of AI and political science. The dataset is available in the supplementary materials.

Downloads

Next from AAAI 2026

D²-VPR: A Parameter-efficient Visual-foundation-model-based Visual Place Recognition Method via Knowledge Distillation and Deformable Aggregation

Similar lecture

Failures to Surface Harmful Contents in Video Large Language Models