Singapore

Interpretability methods have recently gained significant attention, particularly in the context of large language models, enabling insights into linguistic representations, error detection, and model behaviors such as hallucinations and repetitions. However, these techniques remain underexplored in automatic speech recognition (ASR), despite their potential to advance both the performance and interpretability of ASR systems. In this work, we adapt and systematically apply established interpretability methods such as logit lens, linear probing, and activation patching, to examine how acoustic and semantic information evolves across layers in ASR systems. Our experiments reveal previously unknown internal dynamics, including specific encoder-decoder interactions responsible for repetition hallucinations and semantic biases encoded deep within acoustic representations. These insights demonstrate the benefits of extending and applying interpretability techniques to speech recognition, opening promising directions for future research on improving model transparency and robustness.

AAAI 2026

Beyond Transcription: Mechanistic Interpretability in ASR

lalm

mechanistic interpretability

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community." 

*Message from the General Chair, Dirk Hovy*

EMNLP 2025

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

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.

My research focuses on making LLM training and serving
widely accessible to academia and smaller organizations by
reducing dependence on proprietary data and heavy compute.
In this talk, I will present a coherent framework that
unifies data curation without labels, zero-data
self-evolution, reward calibration for reliable confidence
estimation, and computation-efficient language model
inference.

Breaking the Resource Monopoly: LLM Post-Training and Serving with Modest Data and Compute

Machine learning models are often used to make predictions about the outcomes of applications to selective programs. Many prospective school or college applicants turn to machine learning models to predict whether they will be admitted to a program, and employers may use algorithmic tools to filter out resumes predicted to have a low probability of being hired when offering interviews for a job opening. However, such decision processes differ substantially from the conventional machine learning setting: decisions are not independent across applicants. Whether a student is admitted depends on the other applicants who apply because admissions decisions are \textit{capacity-constrained}. We formalize how the nature of admission decisions results in a data-generating process which is incompatible with traditional machine learning assumptions. We characterize how selection functions properties affect the difficulty of generalization to applicant pool distribution shifts, introducing two concepts: stability, which measures how many existing decisions can change when a single new applicant is introduced; and variability, which measures the number of unique students whose decisions can change. We demonstrate our theory on admissions data from the New York City high school matching system, showing that machine learning performance degrades as the applicant pool increasingly differs from the training data. Furthermore, there are larger performance drops for schools using decision rules that are less stable and more variable. Our work raises questions about the reliability of predicting individual admissions probabilities.

Capacity Constraints Make Admissions Processes Less Predictable

As artificial intelligence (AI) becomes increasingly
integrated into daily life, there is a critical need for
developing AI literacy across all educational levels.
However, current AI education remains largely confined to
college-level computer science classrooms with limited
access for K-12 learners. We present the AI Scholars
Program, a novel approach that addresses the AI education
gap by preparing college computing students to serve as AI
education ambassadors in their communities and empowering
K-12 teachers to adopt AI education practices in their
classrooms. This experience report presents the curriculum
and its outcomes after one round of refinement. The program
offers structured AI learning through bi-weekly webinars,
resources, and collaborative opportunities to form teams
and conduct community outreach projects. Our program
invited 63 scholars from 30 institutions across the U.S.,
including 51 college students and 12 K-12 teachers. Their
outreach impacted over 230 K-12 learners. We examine
program outcomes for participants and projects through
pre/post surveys measuring computing attitudes and
self-efficacy for teaching AI, scholar interviews, and
outreach project reports. We share lessons learned and
challenges for designing similar programs, highlighting the
importance of involving educators for effective
community-engaged AI education. The program creates a
sustainable pipeline for college students to develop
technical skills and leadership while addressing K-12 AI
education shortages. We contribute insights for scaling AI
literacy and broadening participation in computing.

AI Scholars Program: Scaling AI Literacy Through K-12 Outreach

Diffusion models have advanced fine-grained garment generation, yet balancing controllability, efficiency, and texture fidelity remains challenging. Adapter-based methods often yield incoherent details, while full fine-tuning is computationally expensive and prone to overwriting pretrained priors.
To address these limitations, we propose IMAGGarment+, an efficient diffusion framework for controllable and high-quality garment synthesis. It comprises two key modules designed for efficient and attribute-aware conditioning.
First, we introduce an attribute-wise feature extractor (AFE) that disentangles key garment attributes, silhouette, logo, position, and color, into parallel latent streams. Each stream is optimized independently via LoRA, ensuring minimal parameter overhead while retaining expressive capacity.
Second, we develop an attribute-adaptive attention (AA) module to inject attribute-specific cues into the generative process through a selective, layer-wise injection strategy. Specifically, silhouette and color features are injected into early decoder layers to guide structural and appearance formation, while logo features are propagated across all layers to ensure cross-scale consistency.
Extensive experiments on fine-grained garment benchmarks demonstrate that IMAGGarment+ outperforms state-of-the-art baselines with less than 20\% additional parameters, validating its effectiveness and efficiency.

IMAGGarment+: Efficient Attribute-Wise Diffusion for Garment Generation

Automatically solving table reasoning tasks remains challenging due to three main factors: (1) diverse and hierarchical table structures that hinder comprehension, (2) the heavy reliance on complex logical and numerical reasoning—which makes purely text-based methods prone to hallucinations—and (3) the necessity of multi-step processing to handle intricate tasks involving multiple and lengthy tables.
To address these challenges, we introduce TaREx, a novel framework that unifies table representation, integrates code-driven execution, and supports interactive multi-step reasoning. TaREx employs a reinforcement learning-based training pipeline to optimize its reasoning policy for complex tasks.
Experimental results show that TaREx achieves state-of-the-art performance across a wide range of table reasoning benchmarks, both in-domain and out-of-domain. These include fundamental tasks such as table question answering (TQA) and table fact verification (TFV), as well as advanced tabular data analysis tasks. The results highlight TaREx’s effectiveness and scalability in advancing automated table reasoning.

TaREx: Reinforcement Learning for Code-Driven Table Reasoning

Recent advances in large language models (LLMs) have improved reasoning in text and image domains, yet achieving robust video reasoning remains a significant challenge. Existing video benchmarks mainly assess shallow understanding and reasoning and allow models to exploit global context, failing to rigorously evaluate true causal and stepwise reasoning. We present CausalStep, a benchmark designed for explicit stepwise causal reasoning in videos. CausalStep segments videos into causally linked units and enforces a strict stepwise question-answer (QA) protocol, requiring sequential answers and preventing shortcut solutions. Each question includes carefully constructed distractors based on error type taxonomy to ensure diagnostic value. The benchmark features 100 videos across six categories and 1,852 multiple-choice QA pairs. We introduce seven diagnostic metrics for comprehensive evaluation, enabling precise diagnosis of causal reasoning capabilities. Experiments with leading proprietary and open-source models, as well as human baselines, reveal a significant gap between current models and human-level stepwise reasoning. CausalStep provides a rigorous benchmark to drive progress in robust and interpretable video reasoning.

CausalStep: A Benchmark for Explicit Stepwise Causal Reasoning in Videos

Generalized Category Discovery (GCD) aims to classify unlabeled data by leveraging knowledge from labeled categories.
While existing methods have achieved remarkable progress, they often treat images as flat feature sets, neglecting the intrinsic hierarchy: where key objects dominate meaning and backgrounds serve as context. For instance, in images of a dog either standing on grass or lying on a bed, the dog remains the central semantic element, whereas the background varies. Motivated by this, we propose LEArning Intrinsic Hierarchy (LEAH), a lightweight plug-and-play module designed to model hierarchical structure within images. LEAH consists of two components: a pruner that filters task-irrelevant tokens to extract key objects, and a constructor that embeds key objects and full images into hyperbolic space using adaptive entailment cones to capture compositional semantics. LEAH can be easily integrated into existing GCD frameworks with minimal modification. When applied to SimGCD, it achieves up to 13.2\% accuracy improvement on fine-grained benchmarks, demonstrating its effectiveness in discovering subtle inter-class differences through hierarchical modeling.

Learning Intrinsic Hierarchy for Generalized Category Discovery

Standpoint logics offer unified modal logic-based formalisms for representing multiple heterogeneous viewpoints. At the same time, many non-monotonic reasoning frameworks can be naturally captured using modal logics — in particular using the modal logic S4F. In this work, we propose a novel formalism called S4F Standpoint Logic, which generalises both S4F and standpoint propositional logic and is therefore capable of expressing multi-viewpoint, non-monotonic semantic commitments. We define its syntax and semantics and analyze its computational complexity, obtaining the result that S4F Standpoint Logic is not computationally harder than its constituent logics, whether in monotonic or non-monotonic form. We also outline mechanisms for credulous and sceptical acceptance and illustrate the framework with an example.

Non-Monotonic S4F Standpoint Logic

Weakly supervised semantic segmentation (WSSS) suffers from an inherent mismatch between coarse image-level annotations and dense pixel-level predictions. To bridge this gap, existing methods primarily focus on generating refined class activation maps (CAM) as pseudo-labels. However, we argue that this focus is insufficient as it overlooks a critical component: the segmentation decoder. The decoder is typically trained through superficial alignment of predictions with pseudo-labels in the logit space. Given the noisy nature of such labels, this naive supervision leads to error accumulation and limits performance. To address this, we propose an Uncertainty-Guided Reliable Learning (UGRL) framework that exerts dual control to reshape the learning process, achieving robust supervision that escapes the CAM shadow. The cornerstone of UGRL is a prototype-driven uncertainty modeling module that estimates the reliability of class-wise supervision. The modeled uncertainty enables two synergistic control mechanisms. First, it adaptively modulates classification and segmentation losses, encouraging the model to learn from more trustworthy signals. Second, it guides the structuring of the decoder’s feature space. Rather than relying solely on superficial alignment, UGRL enforces deeper representation alignment by applying contrastive learning on reliable pixels. This enables rich semantic transfer to fine-grained segmentation details. Extensive experiments on PASCAL VOC and MS COCO demonstrate that our method surpasses other state-of-the-art WSSS methods. Code is available at https://anonymous.4open.science/r/UGRL-70C3/.

Escaping the CAM Shadow: Uncertainty-Guided Reliable Learning for Weakly Supervised Semantic Segmentation

This article proposes a Learning-from-Demonstration (LfD)
method using probability densities on the workspaces of
robot manipulators. The method, named
PRobabilistically-Informed Motion Primitives (PRIMP),
learns the probability distribution of the end effector
trajectories in the 6-D workspace that includes both
positions and orientations. It is able to adapt to new
situations such as novel via points with uncertainty and a
change of viewing frame. The method itself is
robot-agnostic, in that the learned distribution can be
transferred to another robot with the adaptation to its
workspace density. Workspace-STOMP, a new version of the
existing STOMP motion planner, is also introduced, which
can be used as a postprocess to improve the performance of
PRIMP and any other reachability-based LfD method. The
combination of PRIMP and Workspace-STOMP can further help
the robot avoid novel obstacles that are not present during
the demonstration process. The proposed methods are
evaluated with several sets of benchmark experiments. PRIMP
runs more than five times faster than existing
state-of-the-art methods while generalizing trajectories
more than twice as close to both the demonstrations and
novel desired poses. They are then combined with our lab's
robot imagination method that learns object affordances,
illustrating the applicability to learn tool use through
physical experiments.

Downloads

Next from AAAI 2026

Breaking the Resource Monopoly: LLM Post-Training and Serving with Modest Data and Compute

Similar lecture

ASR-EC Benchmark: Evaluating Large Language Models on Chinese ASR Error Correction