United States

Recent advancements in Large Reasoning Models (LRMs) have showcased strong performance across various reasoning tasks by leveraging System‑2 thinking capabilities. However, existing studies indicate that this reasoning ability alone does not reliably transfer to the general alignment domain. Inspired by cognitive science and how humans solve tasks, we argue that LRMs must be equipped with metacognitive knowledge to fully utilize their System‑2 capabilities. In this paper, we propose Priority‑Aware Metacognition (PAM), which guides the model to first identify the top‑level human preference (e.g., harmlessness) as a means of understanding the alignment task’s nature, and then apply other kinds of metacognitive knowledge to better monitor and regulate the model&#39;s thinking process. We implement PAM via a two-stage pipeline: a cold-start phase that collects structured metacognitive knowledge based on Flavell’s theoretical framework, and a preference-optimization phase that further reinforces such metacognition. Extensive experiments validate the effectiveness of PAM. Under the same training pipelines, PAM consistently yields higher performance, improving general domain alignment performance by \textasciitilde 10 points on the helpfulness and harmless benchmarks. Code is available at https://anonymous.4open.science/r/PAM-RM-02DF.

ACL 2026

PAM: Enhancing General Alignment of Large Reasoning Models through Priority-Aware Metacognition

We are pleased to announce the Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI-25), which will be held in Philadelphia, Pennsylvania at the Pennsylvania Convention Center from February 25 to March 4, 2025.

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-25 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.

### [Invited Speakers](https://aaai.org/conference/aaai/aaai-25/aaai-25-invited-speakers/)

Register [here](https://aaai.org/conference/aaai/aaai-25/registration/)

AAAI 2025

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-25 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.



Recent advancements in Large Reasoning Models (LRMs) have showcased strong performance across various reasoning tasks by leveraging System‑2 thinking capabilities. However, existing studies indicate that this reasoning ability alone does not reliably transfer to the general alignment domain. Inspired by cognitive science and how humans solve tasks, we argue that LRMs must be equipped with metacognitive knowledge to fully utilize their System‑2 capabilities. In this paper, we propose Priority‑Aware Metacognition (PAM), which guides the model to first identify the top‑level human preference (e.g., harmlessness) as a means of understanding the alignment task’s nature, and then apply other kinds of metacognitive knowledge to better monitor and regulate the model's thinking process. We implement PAM via a two-stage pipeline: a cold-start phase that collects structured metacognitive knowledge based on Flavell’s theoretical framework, and a preference-optimization phase that further reinforces such metacognition. Extensive experiments validate the effectiveness of PAM. Under the same training pipelines, PAM consistently yields higher performance, improving general domain alignment performance by \textasciitilde 10 points on the helpfulness and harmless benchmarks. Code is available at https://anonymous.4open.science/r/PAM-RM-02DF.

technical paper

It is my great pleasure and honor to welcome you to the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), held in San Diego, California, USA, July 2–7, 2026. ACL 2026 continues our fields tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.

It is exciting to see the many developments in the field reflected at our conference, such as the advancement of large language models (born out of our community) into many application directions. This is not without challenges, partly due to unsolved problems, and partly due to misuse of the technology. All this is reflected well at ACL 2026. The conference also reflects that the field is not only about advancing technology, but also the science of fundamentally understanding language and how it is use by many different speaker communities. 
[Full Message from General Chair Philipp Koehn](https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing) 
<html><button style="display: inline-flex; align-items: center; justify-content: center; white-space: nowrap; border-radius: 9999px; font-weight: bold; background: #7c3aed; color: white; font-family: 'Space Grotesk', sans-serif; height: 40px; font-size: 16px; padding: 0 20px; border: none; cursor: pointer" onclick="window.open('https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing')">Conference Handbook</button></html>

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 ACL 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://2026.aclweb.org/registration/) first.

Registration Required

The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026) will take place in San Diego, California, from July 2 - 7, 2026.

Fine-tuning-as-a-service, while commercially successful for Large Language Model (LLM) providers, exposes models to harmful fine-tuning attacks. As a widely explored defense paradigm against such attacks, unlearning attempts to remove malicious knowledge from LLMs, thereby essentially preventing them from being used to perform malicious tasks. However, we highlight a critical flaw: the inherent general adaptability of LLMs allows them to easily bypass selective unlearning by rapidly relearning or repurposing their general capabilities for harmful tasks. To address this fundamental limitation, we propose a paradigm shift: instead of selective removal, we advocate for inducing model collapse, effectively forcing the model to ''unlearn everything'', specifically in response to updates characteristic of malicious adaptation. This collapse directly neutralizes the very general capabilities that attackers exploit, tackling the core issue unaddressed by selective unlearning. We introduce the Collapse Trap (CTRAP) as a practical mechanism to implement this concept conditionally. Embedded during alignment, CTRAP pre-configures the model's reaction to subsequent fine-tuning dynamics. If updates during fine-tuning constitute a persistent attempt to reverse safety alignment, the pre-configured trap triggers a progressive degradation of the model's core language modeling abilities, ultimately rendering it inert and useless for the attacker. Crucially, this collapse mechanism remains dormant during benign fine-tuning, ensuring the model's utility and general capabilities are preserved.

CTRAP: Embedding Collapse Trap to Safeguard Large Language Models from Harmful Fine-Tuning

Multimodal Continual Instruction Tuning (MCIT) is essential for adapting Multimodal Large Language Models (MLLMs) to dynamic data streams, yet preventing catastrophic forgetting remains a major challenge. Existing parameter-efficient approaches often face a dilemma: fixed architectures suffer from knowledge interference, while dynamic strategies incur inefficient capacity expansion, limiting scalability. We propose MoBLoRA (Mixture-of-Bases LoRA), a novel framework for MCIT. Motivated by our geometric analysis revealing subspace redundancy across sequential tasks, MoBLoRA shifts the paradigm from expert selection to subspace mixing: it decomposes adaptation weights into a globally shared pool of orthonormal bases to capture task-invariant knowledge, and lightweight mixing matrices to encode task-specific variations. This design effectively decouples knowledge accumulation from task reconstruction. Experiments on standard benchmarks show MoBLoRA significantly outperforms state-of-the-art methods while maintaining superior parameter efficiency.

From Experts to Bases: Orthogonal Subspace Mixture for Continual Multimodal Instruction Tuning

Generating SVG-based fonts requires Multimodal Large Language Models (MLLMs) to translate high-level linguistic intent into low-level, topologically constrained symbolic sequences. However, current approaches struggle with two fundamental misalignments: the semantic ambiguity of unstructured natural language for precise geometric control, and the inefficiency of generic text tokenizers, which fragment coordinate-dense SVG XML into excessively long sequences with low information density. In this work, we propose Vector Calligrapher, a system that treats SVG generation as a conditional language modeling task optimized for both semantic grounding and representational efficiency.

To bridge the semantic gap, we introduce a structured linguistic supervision Font Description Framework that decomposes typographic style into interpretable linguistic dimensions (e.g., historical lineage, affective metaphors), providing structured supervision aligned with the compositional syntax of SVG. To address the tokenization bottleneck, we design a scalable separated-coordinate strategy that bypasses the vocabulary explosion of flattened tokens while significantly compressing sequence length. Supported by VectorFont, a dataset of over 10 million hierarchically annotated glyphs, our approach improves CLIP score by +23%, reduces geometric error by ≈48%, and boosts generation efficiency by achieving an 18% Command-per-Token (C/T) ratio---a 6× increase in information density over standard baselines. These results demonstrate that combining structured linguistic supervision with efficient symbolic tokenization is essential for reliable, controllable vector graphics synthesis. VectorFont dataset, Code and model weights will be publicly released.

Vector Calligrapher: Generating Scalable Vector Graphics via Structured Linguistic Supervision

As Large Language Model (LLM) agents increasingly leverage the Model Context Protocol (MCP) to operate in complex environments, the expansion of their action spaces offers agents unsafe capabilities and underscores the risk of power-seeking. While broad action space and greater environment influence are essential for task fulfillment, they creates a fragile risk surface where minor errors or hallucinations are magnified into catastrophic failures. In response, we propose SafeMCP, a server-side defense plugin that constrains tool acquisition via predictive reasoning regarding future safety risks. SafeMCP utilizes an internal world model for look-ahead reasoning to implement a two-tier defense: proactive tool filtering to constrain hazardous power expansion and immediate intervention as a fail-safe. To train SafeMCP, we introduce a three-stage pipeline comprising environmental dynamic grounding, safe policy initialization, and reinforcement learning (RL) with dual verifiable rewards. Experiments on PowerSeeking Bench, ToolEmu, and AgentHarm show that SafeMCP achieves a safe equilibrium, effectively mitigating risks while preserving agent utility.

SafeMCP: Proactive Power Regulation for LLM Agent Defense via Environment-Grounded Look-Ahead Reasoning

Role-playing agents (RPAs) require balancing multiple objectives, such as instruction following, persona consistency, and stylistic fidelity, which are not always perfectly aligned across different dimensions. While prior work has primarily relied on supervised fine-tuning or reinforcement learning with scalarized rewards, these approaches do not explicitly address the coordination of multiple reward dimensions during optimization. We present **MOA** (**M**ulti-**O**bjective **A**lignment), a reinforcement-learning framework that enables multi-dimensional, fine-grained rubric optimization for general RPAs. MOA introduces a novel multi-objective optimization strategy that trains simultaneously on multiple fine-grained rubrics to boost optimization performance. Besides, to address the issues of model output diversity and quality, we have also employed thought-augmented rollout with off-policy guidance. Experiments on PersonaGym and RoleMRC show that MOA consistently improves multi-dimensional role-playing performance over supervised and standard RL baselines. Under identical evaluation protocols, an 8B model trained with MOA reaches performance competitive with strong closed-source models across multiple evaluation dimensions. These results suggest that MOA provides a practical framework for training more capable general-purpose role-playing agents.

MOA: Multi-Objective Alignment for Role-Playing Agents

While LLMs are powerful embedding backbones, their application in training-free settings faces two structural challenges: causal attention restricts early tokens from accessing subsequent context, and the next-token prediction objective biases representations toward generation rather than semantic compression. To address these limitations, we propose KV-Embedding, a framework that activates the latent representation power of frozen LLMs. Our method leverages the observation that the key-value (KV) states of the final token at each layer encode a compressed view of the sequence. By re-routing these states as a prepended prefix, we enable all tokens to access sequence-level context within a single forward pass. To ensure model-agnostic applicability, we introduce an automated layer selection strategy based on intrinsic dimensionality. Evaluations on MTEB across Qwen, Mistral, and Llama backbones show that KV-Embedding outperforms existing training-free baselines by up to 10%, while maintaining robust performance on sequences up to 4,096 tokens. These results demonstrate that internal state manipulation offers an efficient alternative to input modification, and we hope this work encourages further exploration of LLM internals for representation learning.

KV-Embedding: Training-free Text Embedding via Internal KV Re-routing in Decoder-only LLMs

Chain-of-thought (CoT) reasoning has emerged as a crucial paradigm for enhancing large language model (LLM) performance on multi-step reasoning tasks.
However, the internal mechanisms by which LLMs invoke knowledge and propagate information across different steps of the CoT are poorly understood.
To fill this gap, we propose a multi-stage probing framework that enforces structured reasoning with three explicit stages: keyword extraction, theorem generation, and computation execution.
The framework integrates attention knockout to trace cross-layer information flow and theorem probing to examine how specific contents are encoded within representations.
To enable controlled and stage-aligned analysis, we construct a structured CoT dataset that covers the mathematics and physics domains. Experiments on four instruction-tuned LLMs reveal distinct stage-specific patterns.
First, keyword information is progressively aggregated into the final token in later layers.
Second, theorem semantics are encoded in the mid-to-late layers and undergo two stages of propagation.
Finally, parameter substitution is achieved through joint extraction by the final token and other tokens.
The first parameter predominantly relies on the final token, whereas later parameters increasingly depend on information extracted by other tokens.
Overall, our findings shed light on the neural implementation of CoT reasoning and provide actionable insights for developing more interpretable and reasoning-capable LLMs.
We further evaluate a free-form prompting setting without labeled fields and observe consistent qualitative trends.

Exploring Layer Activation Dynamic of CoT via Knowledge Probe

Unified Information Extraction (UIE) aims to handle heterogeneous IE tasks within a single framework, but existing methods often suffer from inconsistent schema representation, implicitly intermediate reasoning and full-parameter adaptation, which limit generalization, interpretability and parameter efficiency. To address these issues, we propose UC-UIE (Universal Capabilities-based Unified Information Extractor), a unified framework based on Large Language Model (LLM), which introduces a unified frame-and-slots schema for IE tasks and explicitly decomposes IE reasoning into three universal capabilities: judging, locating, and associating. Furthermore, UC-UIE adopts a Low-Rank Adaptation (LoRA) based hierarchical Mixture-of-Experts (MoE) adapter to fine-tune LLMs for IE tasks, which explicitly models these three capabilities in a task-driven way while ensuring parameter efficiency. With only 1.24% trainable parameters, UC-UIE outperforms full-parameter tuning methods, showing excellent parameter efficiency. Zero-shot evaluation reveals its strong generalization ability to unseen domains and schemas, benefiting from unified schema representation and explicit capability decomposition. Further experiments validate that the hierarchical MoE adapter learns capability specialization and composition, which enhances both UIE performance and interpretability.

Capability Decomposition for Unified Information Extraction via Hierarchical Mixture-of-Experts

The widespread utilization of language models in modern applications is inconceivable without Parameter-Efficient Fine-Tuning techniques, such as low-rank adaptation ($\texttt{LoRA}$), which adds trainable adapters to selected layers. Although $\texttt{LoRA}$ may obtain accurate solutions, it requires significant memory to train large models and intuition on which layers to add adapters. In this paper, we propose a novel method, $\texttt{WeightLoRA}$, which overcomes this issue by adaptive selection of the most critical $\texttt{LoRA}$ heads throughout the optimization process. As a result, we can significantly reduce the number of trainable parameters while maintaining the capability to obtain consistent or even superior metric values. We conduct experiments for a series of competitive benchmarks and DeBERTa, BART, Llama and Qwen models, comparing our method with different adaptive approaches. The experimental results demonstrate the efficacy of $\texttt{WeightLoRA}$ and the superior performance of $\texttt{WeightLoRA+}$ in almost all cases. The source code is available at https://github.com/brain-lab-research/WLoRA

WeightLoRA: Keep Only Necessary Adapters

LLM-based agents are increasingly deployed to autonomously solve complex tasks, raising urgent needs for IP protection and regulatory provenance. While content watermarking effectively attributes LLM-generated outputs, it fails to directly identify the high-level planning behaviors (e.g., tool and subgoal choices) that govern multi-step execution. Critically, watermarking at the planning-behavior layer faces unique challenges: minor distributional deviations in decision-making can compound during long-term agent operation, degrading utility, and many agents operate as black boxes that are difficult to intervene in directly. To bridge this gap, we propose AgentMark, a behavioral watermarking framework that embeds multi-bit identifiers into planning decisions while preserving utility. It operates by eliciting an explicit behavior distribution from the agent and applying distribution-preserving conditional sampling, enabling deployment under black-box APIs while remaining compatible with action-layer content watermarking. Experiments across embodied, tool-use, and social environments demonstrate practical multi-bit capacity, robust recovery from partial logs, and utility preservation. Code is available at https://github.com/Tooooa/AgentMark.

Downloads

Next from ACL 2026

CTRAP: Embedding Collapse Trap to Safeguard Large Language Models from Harmful Fine-Tuning

Similar lecture

Decoupling Metacognition from Cognition: A Framework for Quantifying Metacognitive Ability in LLMs