Singapore

Large language models have gained widespread attention recently, but their potential security vulnerabilities, especially privacy leakage, are also becoming apparent. To test and evaluate for data extraction risks in LLM, we proposed CoSPED, short for Consistent Soft Prompt targeted data Extraction and Defense. We introduce several innovative components, including Dynamic Loss, Additive Loss, Common Loss, and Self Consistency Decoding Strategy, and tested to enhance the consistency of the soft prompt tuning process. Through extensive experimentation with various combinations, we achieved an extraction rate of 65.2% at a 50-token prefix comparison. Our comparisons of CoSPED with other reference works confirm our superior extraction rates. We evaluate CoSPED on more scenarios, achieving Pythia model extraction rate of 51.7% and introducing cross-model comparison. Finally, we explore defense through Rank-One Model Editing and achieve a reduction in the extraction rate to 1.6%, which proves that our analysis of extraction mechanisms can directly inform effective mitigation strategies against soft prompt-based attacks.

AAAI 2026

CoSPED: Consistent Soft Prompt Targeted Data Extraction and Defense

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.

In vision-language models (VLMs), the ability to perceive and interpret color and physical environment is crucial for achieving contextually accurate understanding and interaction. However, despite advances in multimodal modeling, there remains a significant lack of specialized datasets that rigorously evaluate a model's capacity to discern subtle color variations and spatial context---critical elements for situational comprehension and reliable deployment across real-world applications.
Toward that goal, we curate MegaCoin, a high-quality, human-labeled dataset based on \emph{real} images with various contextual attributes. MegaCoin consists of two parts: MegaCoin-Instruct, which serves as a supervised fine-tuning (SFT) dataset for VLMs; and MegaCoin-Bench, an annotated test set that can be used as a stand-alone QA dataset. MegaCoin provides three annotated features for 220,000 real images: foreground color, background color, and description of an object's physical environment, constituting 660k human annotations. In addition, MegaCoin can be applied to benchmark domain generalization (DG) algorithms. We explore benchmarking DG methods in the linear probing setup for VLM and show some new insights. Last but not least, we show that VLMs, including GPT-4o, have subpar color recognition capabilities, and fine-tuning with MegaCoin can result in improved performance on visual evaluation tasks. In certain cases, MegaCoin fine-tuned small-scale open-source models such as LLaVA and Bunny can outperform closed-source GPT-4o. We hope the utilities of MegaCoin can shed light on the directions VLMs can improve and provide a more complex platform for domain generalization algorithms.

MegaCoin: Enhancing Medium-Grained Color Perception for Vision-Language Models

The increasing complexity of modern AI systems exposes a significant assurance gap: safety evidence from practices like red-teaming and robustness testing remains fragmented, lacking a formal mechanism for composition and propagation throughout the development lifecycle. This prevents the construction of rigorous, dynamic safety cases essential for trustworthy AI. We introduce the Composable Assurance Framework (CAF), a novel engineering methodology that integrates safety assurance directly into MLOps workflows. At its core is the Formal Safety Assertion (FSA), a standardized, machine-readable structure that verifiably links safety properties—such as robustness scores or the absence of deceptive circuits—to specific AI artifacts. We then define a Composition Calculus, a set of formal rules governing how FSAs are propagated and aggregated as components are combined into a system. This approach transforms the development pipeline into an automated evidence-gathering engine, whose output is a dynamic Directed Acyclic Graph (DAG) of assertions that constitutes a living safety case. Through a prototype and a Retrieval-Augmented Generation (RAG) case study, we demonstrate how CAF automatically enforces a predefined safety policy, blocking non-compliant deployments.

Composable Assurance for AI Alignment: A Framework for Propagating Formal Safety Properties Through MLOps

Large language models are widely used, but aligning them with societal values remains challenging. Current approaches often rely on human annotations, which are hard to scale, or synthetic data produced by models that may themselves be misaligned, making it difficult to capture genuine public opinion. This limits scalability and introduces demographic biases that reduce the representativeness and fairness of model behavior. We introduce a novel approach to pluralistic alignment through behavioral learning, grounded in the psychological principle that actions (behavior) have strong consistency with opinions. Specifically, we present ALPHA50M, a dataset of over 50 million samples derived from 1.5 million real-world advertisements, incorporating rich behavioral signals inferred from demographic engagement patterns. Models trained on this data achieve state-of-the-art zero-shot performance on diverse alignment benchmarks spanning cultural reasoning, political views, and social values. We also propose two new benchmarks: OpinionQA-XL, which covers surveys across 100+ societal topics, and GSS, which evaluates temporal opinion shift modeling over decades. Our results demonstrate that learning from behavioral signals, derived from observed human actions, enables models to align with diverse demographic opinions, capture underlying social and cultural norms, and generalize to new topics and surveys beyond training data. This behavioral learning paradigm offers a scalable and demographically broad alternative to existing alignment techniques.

ALPHA: Action-Based Learning for Pluralistic Human Alignment in Large Language Models

Vision-language models (VLMs) have made significant strides in reasoning, yet they often struggle with complex multimodal tasks and tend to generate overly verbose outputs. A key limitation is their reliance on chain-of-thought (CoT) reasoning, despite many tasks benefiting from alternative topologies like trees or graphs. To address this, we introduce STELAR-Vision, a training framework for topology-aware reasoning. At its core is TopoAug, a synthetic data pipeline that enriches training with diverse topological structures. Using supervised fine-tuning and reinforcement learning, we post-train Qwen2VL models with both accuracy and efficiency in mind. Additionally, we propose Frugal Learning, which reduces output length with minimal accuracy loss. On MATH-V and VLM S2H, STELAR-Vision improves accuracy by 9.7% over its base model and surpasses the larger Qwen2VL-72B-Instruct by 7.3%. On five out-of-distribution benchmarks, it outperforms Phi-4-Multimodal-Instruct by up to 36% and LLaMA-3.2-11B-Vision-Instruct by up to 13.2%, demonstrating strong generalization. Compared to Chain-Only training, our approach achieves 4.3% higher overall accuracy on in-distribution datasets and consistently outperforms across all OOD benchmarks. The data and code will be available.

STELAR-VISION: Self-Topology-Aware Efficient Learning for Aligned Reasoning in Vision

Language models (LMs) judges are widely used to evaluate the quality of LM outputs. Despite many advantages, LM judges display concerning biases that can impair their integrity in evaluations. One such bias is self-preference: LM judges preferring their own answers over those produced by other LMs or humans. The bias is hard to eliminate as frontier LM judges can distinguish their own outputs from those of others, even when the evaluation candidates are not labeled with their sources. In this paper, we investigate strategies to mitigate self-preference by reducing the LM judges' ability to recognize their own outputs. We apply black-box perturbations to evaluation candidates in pairwise comparison to obfuscate the authorship and reduce self-recognition. We find that perturbations as simple as synonym replacement for a few words predictably reduce self-preference. However, we also uncover fundamental challenges to eliminating the bias: when we extrapolate our perturbations to a more complete neutralization of stylistic differences between the evaluation candidates, self-preference recovers. Our findings suggest that self-recognition and self-preference can happen on many semantic levels, and a complete mitigation remains challenging despite promising initial results.

Mitigating Self-Preference by Authorship Obfuscation

Large language models (LLMs) have been shown to exhibit social bias however, bias towards non-protected stigmatized identities remain understudied. Furthermore, what social features of stigmas are associated with bias in LLM outputs is unknown. From psychology literature, it has been shown that stigmas contain six shared features: aesthetics, concealability, course, disruptiveness, origin, and peril. In this study, we investigate if human and LLM ratings of the features of stigmas, along with prompt style and type of stigma, have effect on bias towards stigmatized groups in LLM outputs. We measure bias against 93 stigmatized groups across three widely used LLMs (Granite 3.0-8B, Llama-3.1-8B, Mistral-7B) using SocialStigmaQA, a benchmark that includes 37 social scenarios about stigmatized identities; for example deciding whether to recommend them for an internship. We find that stigmas rated by humans to be highly perilous (e.g., being a gang member or having HIV) have the most biased outputs from SocialStigmaQA prompts (60% of outputs from all models) while sociodemographic stigmas (e.g. Asian- American or old age) have the least amount of biased outputs (11%). We test if the amount of biased outputs could be decreased by using guardrail models, models meant to identify harmful input, using each LLM’s respective guardrail model (Granite Guardian 3.0, Llama Guard 3.0, Mistral Moderation API). We find that bias decreases significantly by 10.4% , 1.4%, and 7.8%, respectively. However, we show that features with significant effect on bias remain unchanged post-mitigation and that guardrail models often fail to recognize the intent of bias in prompts. This work has implications for using LLMs in scenarios involving stigmatized groups and we suggest future work towards improving guardrail models for bias mitigation.

Identifying Features Associated with Bias Against 93 Stigmatized Groups in Language Models and Guardrail Model Safety Mitigation

The existence of multiple, equally accurate models for a given predictive task—the Rashomon set—leads to predictive multiplicity, where models achieve similar global accuracy but diverge in their individual predictions. This inconsistency undermines trust in high-stakes applications, where reliable decision-making at the individual level is critical. Existing reconciliation methods attempt to resolve this issue by enforcing global agreement across models, but often fall short in ensuring consistency for specific instances. We first introduce Rocile (Robust Reconciliation), a model-agnostic framework that systematically reduces ensemble disagreement by iteratively adjusting model predictions through a momentum-based batching procedure. Rocile guarantees convergence to a globally consistent ensemble, yet, like other global methods, it overlooks localized inconsistencies that impact individual predictions. To address this limitation, we propose AdaRocile, an extension of Rocile that incorporates local calibration into the reconciliation process. For each test instance, AdaRocile identifies a context-sensitive neighborhood using an adaptive nearest-neighbor strategy, computes empirical correction terms for each model based on residuals in the local neighborhood, and applies Rocile to reconcile the locally adjusted predictions into a globally coherent output. The reconciled predictions are then distilled into a single, transparent decision rule for real-world deployment. Empirical results on high-stakes benchmarks, such as COMPAS and Adult, show that AdaRocile significantly improves the accuracy-reliability trade-off, reducing local calibration error by up to 27.1% over the global-only reconciliation, while driving key multiplicity metrics (variance, ambiguity, discrepancy, and disagreement rate) to near zero, maintaining global performance on key benchmarks. AdaRocile delivers both interpretability and individual-level reliability, offering a scalable, practical pipeline for building trustworthy and aligned AI systems.

Resolving Predictive Multiplicity for the Rashomon Set

Beyond simple text generation, Large Language Models (LLMs) have evolved into agentic systems capable of planning and interacting with external tools to solve complex tasks. This evolution involves fine-tuning LLMs on agent-specific tasks to enhance their proficiency. However, safety concerns are frequently overlooked during this fine-tuning process. In this work, we show that aligned LLMs can become unintentionally misaligned, leading to a higher likelihood of executing harmful tasks and a reduced tendency to refuse them when fine-tuned to execute agentic tasks. To address these safety challenges, we propose Prefix INjection Guard (PING), a simple yet effective method that prepends automatically generated natural language prefixes to agent responses, guiding them to refuse harmful requests while preserving performance on benign tasks. Specifically, we introduce an iterative approach that alternates between (1) generating candidate prefixes and (2) selecting those that optimize both task performance and refusal behavior. Experimental results demonstrate that PING significantly enhances the safety of fine-tuned LLM agents without sacrificing their effectiveness. PING consistently outperforms existing prompting approaches across diverse benchmarks in both web navigation and code generation tasks. Our analysis of internal hidden states via linear probes reveals that prefix tokens are crucial for behavior modification, explaining the performance gains. WARNING: This paper contains contents that are unethical or offensive in nature.

Unintended Misalignment from Agentic Fine-Tuning: Risks and Mitigation

Large language models require consistent behavioral patterns for safe deployment, yet their personality-like traits remain poorly understood. We present PERSIST (PERsonality Stability in Synthetic Text), a comprehensive evaluation framework testing 25+ open-source models (1B-685B parameters) across 500,000+ responses. Using traditional (BFI-44, SD3) and novel LLM-adapted personality instruments, we systematically vary question order, paraphrasing, personas, and reasoning modes.
Our findings challenge fundamental deployment assumptions: (1) Even 400B+ models exhibit substantial response variability (SD > 0.4); (2) Minor prompt reordering alone shifts personality measurements by up to 20%; (3) Interventions expected to stabilize behavior, such as chain-of-thought reasoning, detailed personas instruction, inclusion of conversation history, can paradoxically increase variability; (4) LLM-adapted instruments show equal instability to human-centric versions, confirming architectural rather than translational limitations.
This persistent instability across scales and mitigation strategies suggests current LLMs lack the foundations for genuine behavioral consistency. For safety-critical applications requiring predictable behavior, these findings indicate that personality-based alignment strategies may be fundamentally inadequate.

Persistent Instability in LLM’s Personality Measurements: Effects of Scale, Reasoning, and Conversation History

Large Language Models (LLMs) are prone to generating fluent but incorrect content, known as confabulation, which poses increasing risks in multi-turn or agentic applications where outputs may be reused as context. In this work, we investigate how in-context information influences model behavior and whether LLMs can identify their unreliable responses. We propose a reliability estimation that leverages token-level uncertainty to guide the aggregation of internal model representations. Specifically, we compute aleatoric and epistemic uncertainty from output logits to identify salient tokens and aggregate their hidden states into compact representations for response-level reliability prediction. Through controlled experiments on open QA benchmarks, we find that correct in-context information improves both answer accuracy and model confidence, while misleading context often induces confidently incorrect responses, revealing a misalignment between uncertainty and correctness. Our probing-based method captures these shifts in model behavior and improves the detection of unreliable outputs across multiple open-source LLMs. These results underscore the limitations of direct uncertainty signals and highlight the potential of uncertainty-guided probing for reliability-aware generation.

Content not yet available

Downloads

Next from AAAI 2026

MegaCoin: Enhancing Medium-Grained Color Perception for Vision-Language Models

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Content not yet available

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AAAI 2026

MegaCoin: Enhancing Medium-Grained Color Perception for Vision-Language Models

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads