Jerry Huang

Beyond Hard Writes and Rigid Preservation: Soft Recursive Least-Squares for Lifelong LLM Editing

Xinyu Wang

Yu Gu

Peng Lu

Yufei Cui

Xiao-Wen Chang

Model editing updates a pre-trained LLM with new facts or rules without retraining while preserving unrelated behavior. In real deployment, … (voir plus)edits arrive as long streams, creating a plasticity-stability dilemma: repeated locate-then-edit "hard writes" can accumulate interference over time, while rigid preservation constraints may protect only explicitly constrained directions, allowing past edits or unconstrained behaviors to deviate. We propose RLSEdit, a recursive least-squares editor for long sequential editing. RLSEdit formulates editing as an online quadratic optimization with soft constraints, minimizing a cumulative key-value fitting objective together with two regularizers that control deviation from the pre-trained weights and from a designated anchor mapping. This objective admits an efficient Woodbury-based online recursion, with per-edit cost independent of history length and scaling only with the current edit size. We further provide deviation bounds and an asymptotic characterization of the adherence-preservation trade-off in the many-edits regime. Experiments on CounterFact and ZsRE across multiple model families show stable scaling to 10K edits, outperforming strong baselines in both edit success and holistic stability, while retaining early edits and preserving general capabilities on GLUE and held-out reasoning/code benchmarks.

2026-08-31

International Joint Conference on Artificial Intelligence (publié)

doi.org

arxiv.org

FinSAgent: Corpus-Aligned Multi-Agent RAG Framework for Evidence-Grounded SEC Filing Question Answering

Jijun Chi

Zhenghan Tai

Hanwei Wu

Tung Sum Thomas Kwok

Hailin He

Zixing Liao

Bohuai Xiao

Chaolong Jiang

Jianliang Lei

Jerry Huang

Peng Lu

Muzhi Li

Liheng Ma

Yihong Wu

Sicheng Lyu

Jingrui Tian

Yihan Li

Yanzhang Ma

Sizhe Guan

Dingtao Hu … (voir 4 de plus)

Yufei Cui

Ling Zhou

Lei Ding

Xinyu Wang

Financial question answering over U.S. Securities and Exchange Commission (SEC) filings requires retrieving and synthesizing heterogeneous e… (voir plus)vidence dispersed across long, standardized, and highly redundant disclosures. Existing retrieval-augmented and multi-agent systems typically derive retrieval queries directly from the user's question and rank candidates by semantic similarity. Together, these choices create prior-corpus misalignment: a mismatch between model priors and the target filings' structure, terminology, and evidence standards. As a result, query generation misses corpus-specific evidence, while semantic reranking favors topically similar but evidentially invalid false-positive chunks. We propose FinSAgent, an evidence-grounded multi-agent framework that reframes SEC filing QA as corpus-aligned retrieval planning and corrects both ends with a single principle: inject corpus-side conditioning wherever model priors would otherwise dominate. FinSAgent combines (1) role-specialized agents anchored to the mandated 10-K item structure, (2) database-aware query decomposition that conditions each agent's sub-queries on a lightweight, summary-level view of the local corpus, and (3) multi-path retrieval with a learned feature-gated reranker that separates evidential validity from semantic similarity. Across five offline financial QA benchmarks, FinSAgent improves retrieval coverage and answer correctness over strong single-agent and multi-agent baselines; in a three-arm randomized online experiment with 1,000 anonymous user ratings, it also receives higher scores than baselines.

2026-07-19

arXiv (prépublication)

doi.org

arxiv.org

Investigating the Multilingual Calibration Effects of Language Model Instruction-Tuning

Jerry Huang

Peng Lu

Qiuhao Zeng

Yusuke Iwasawa

Yutaka Matsuo

A. Chandar

Edison Marrese-Taylor

Irene Li

Ensuring that deep learning models are well-calibrated in terms of their predictive uncertainty is essential in maintaining their trustworth… (voir plus)iness and reliability, yet despite increasing advances in foundation model research, the relationship between such large language models (LLMs) and their calibration remains an open area of research. In this work, we look at a critical gap in the calibration of LLMs within multilingual settings, in an attempt to better understand how the data scarcity can potentially lead to different calibration effects and how commonly used techniques can apply in these settings. Our analysis on two multilingual benchmarks, over 29 and 42 languages respectively, reveals that even in low-resource languages, model confidence can increase significantly after instruction-tuning on high-resource language SFT datasets. However, improvements in accuracy are marginal or non-existent, resulting in mis-calibration, highlighting a critical shortcoming of standard SFT for multilingual languages. Furthermore, we observe that the use of label smoothing to be a reasonable method alleviate this concern, again without any need for low-resource SFT data, maintaining better calibration across all languages. Overall, this highlights the importance of multilingual considerations for both training and tuning LLMs in order to improve their reliability and fairness in downstream use.

2026-01-03

ArXiv (prépublication)

doi.org

arxiv.org

Attention with Routed-Memory for Learnable Sparse Control

Qiuhao Zeng

Jerry Huang

Peng Lu

Ruiyi Fang

Gezheng Xu

Zihao Jing

Yufei Cui

Charles X. Ling

Gang Niu

Boyu Wang

Despite advances in long-context inference, large language models (LLMs) remain fundamentally limited by the key-value (KV) caching mechanis… (voir plus)ms that are necessary for stable computation. Management techniques, such as selective token eviction and pruning, have vastly mitigated the issues that have arisen, but often discard potentially useful information to manage the growing memory requirements of the cache. In this paper, we build upon these approaches to propose Attention with Routed Memory ARM, a novel KV caching structure that introduces a fully differentiable, fixed-size memory system organized as a hierarchical routing structure that learns to select memory slots via Gumbel-Softmax and performs sigmoid-gated updates that softly combine new and stored information, avoiding hard eviction and thereby reducing information loss. By combining this with a policy to dynamically select varying amounts of memory at inference, ARM adapts its accesses for simple contexts and expanding retrieval for inputs that require deeper reasoning, enabling more scalable and effective retrieval on both short and long contexts. Experimental results on standard commonsense and long-context reasoning benchmarks demonstrate that ARM achieves superior performance and efficiency compared to fixed KV-caching approaches, while remaining efficient and scalable in terms of both memory and generation latency.

2025-12-31

International Conference on Machine Learning (Accept (regular))

openreview.net

Calibrated Language Models and How to Find Them with Label Smoothing

Jerry Huang

Peng Lu

Qiuhao Zeng

Recent advances in natural language processing (NLP) have opened up greater opportunities to enable fine-tuned large language models (LLMs) … (voir plus)to behave as more powerful interactive agents through improved instruction-following ability. However, understanding how this impacts confidence calibration for reliable model output has not been researched in full. In this work, we examine various open-sourced LLMs, identifying significant calibration degradation after instruction tuning in each. Seeking a practical solution, we look towards label smoothing, which has been shown as an effective method to regularize for overconfident predictions but has yet to be widely adopted in the supervised fine-tuning (SFT) of LLMs. We first provide insight as to why label smoothing is sufficient to maintain calibration throughout the SFT process. However, settings remain where the effectiveness of smoothing is severely diminished, in particular the case of large vocabulary LLMs (LV-LLMs). We posit the cause to stem from the ability to become over-confident, which has a direct relationship with the hidden size and vocabulary size, and justify this theoretically and experimentally. Finally, we address an outstanding issue regarding the memory footprint of the cross-entropy loss computation in the label smoothed loss setting, designing a customized kernel to dramatically reduce memory consumption without sacrificing speed or performance in comparison to existing solutions for non-smoothed losses.

2025-07-14

International Conference on Machine Learning (Accept (poster))

doi.org

proceedings.mlr.press

Parity Requires Unified Input Dependence and Negative Eigenvalues in SSMs

Jayesh Khullar

Franccois Rivest

A. Chandar

2025-06-09

ICML.cc/2025/Workshop/MOSS (publié)

doi.org

openreview.net

ZETA: Leveraging $Z$-order Curves for Efficient Top-$k$ Attention

Qiuhao Zeng

Jerry Huang

Peng Lu

Gezheng Xu

Boxing Chen

Charles Ling

Boyu Wang

Over recent years, the Transformer has become a fundamental building block for sequence modeling architectures. Yet at its core is the use o… (voir plus)f self-attention, whose memory and computational cost grow quadratically with the sequence length

2025-04-22

International Conference on Learning Representations (Accept (Poster))

openreview.net

Do Robot Snakes Dream like Electric Sheep? Investigating the Effects of Architectural Inductive Biases on Hallucination

Jerry Huang

Prasanna Parthasarathi

Mehdi Rezagholizadeh

Boxing Chen

A. Chandar

The growth in prominence of large language models (LLMs) in everyday life can be largely attributed to their generative abilities, yet some … (voir plus)of this is also owed to the risks and costs associated with their use. On one front is their tendency to \textit{hallucinate} false or misleading information, limiting their reliability. On another is the increasing focus on the computational limitations associated with traditional self-attention based LLMs, which has brought about new alternatives, in particular recurrent models, meant to overcome them. Yet it remains uncommon to consider these two concerns simultaneously. Do changes in architecture exacerbate/alleviate existing concerns about hallucinations? Do they affect how and where they occur? Through an extensive evaluation, we study how these architecture-based inductive biases affect the propensity to hallucinate. While hallucination remains a general phenomenon not limited to specific architectures, the situations in which they occur and the ease with which specific types of hallucinations can be induced can significantly differ based on the model architecture. These findings highlight the need for better understanding both these problems in conjunction with each other, as well as consider how to design more universal techniques for handling hallucinations.

2024-12-31

ACL (Findings) (publié)

doi.org

arxiv.org

Do Large Language Models Know How Much They Know?

Gabriele Prato

Jerry Huang

Prasanna Parthasarathi

Shagun Sodhani

A. Chandar

2024-10-31

Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (publié)

doi.org

arxiv.org

Promoting Exploration in Memory-Augmented Adam using Critical Momenta

Pranshu Malviya

Goncalo Mordido

Aristide Baratin

Reza Babanezhad Harikandeh

Adaptive gradient-based optimizers, notably Adam, have left their mark in training large-scale deep learning models, offering fast convergen… (voir plus)ce and robustness to hyperparameter settings. However, they often struggle with generalization, attributed to their tendency to converge to sharp minima in the loss landscape. To address this, we propose a new memory-augmented version of Adam that encourages exploration towards flatter minima by incorporating a buffer of critical momentum terms during training. This buffer prompts the optimizer to overshoot beyond narrow minima, promoting exploration. Through comprehensive analysis in simple settings, we illustrate the efficacy of our approach in increasing exploration and bias towards flatter minima. We empirically demonstrate that it can improve model performance for image classification on ImageNet and CIFAR10/100, language modelling on Penn Treebank, and online learning tasks on TinyImageNet and 5-dataset. Our code is available at https://github.com/chandar-lab/CMOptimizer.

2024-06-08

Transactions on Machine Learning Research (accepté)

doi.org

openreview.net

Manifold Metric: A Loss Landscape Approach for Predicting Model Performance

Pranshu Malviya

Jerry Huang

Quentin Fournier

A. Chandar

Determining the optimal model for a given task often requires training multiple models from scratch, which becomes impractical as dataset an… (voir plus)d model sizes grow. A more efficient alternative is to expand smaller pre-trained models, but this approach is underutilized due to a limited understanding of its impact on the training dynamics. Existing methods for quantifying this impact have notable limitations, including computation cost. To address this, we introduce a new perspective based on the loss landscape, which has been shown to contain a manifold of linearly connected minima. Specifically, we propose a metric that estimates the size of this manifold to study the impact of model expansion. Our experiments reveal a strong correlation between performance gains and our manifold metric, enabling more informed model comparison and offering a first step toward a geometry-driven approach for reliable model expansion. Notably, our metric outperforms other baselines, even when different types of expansion with equivalent number of parameters are applied to a model.

2024-05-23

ArXiv (prépublication)

arxiv.org

Predicting the Impact of Model Expansion through the Minima Manifold: A Loss Landscape Perspective

The optimal model for a given task is often challenging to determine, requiring training multiple models from scratch which becomes prohibit… (voir plus)ive as dataset and model sizes grow. A more efficient alternative is to reuse smaller pre-trained models by expanding them, however, this is not widely adopted as how this impacts training dynamics remains poorly understood. While prior works have introduced statistics to measure these effects, they remain flawed. To rectify this, we offer a new approach for understanding and quantifying the impact of expansion through the lens of the loss landscape, which has been shown to contain a manifold of linearly connected minima. Building on this new perspective, we propose a metric to study the impact of expansion by estimating the size of the manifold. Experimental results show a clear relationship between gains in performance and manifold size, enabling the comparison of candidate models and presenting a first step towards expanding models more reliably based on geometric properties of the loss landscape.

2024-05-23

arXiv (prépublication)

doi.org

arxiv.org

Communauté de pratique de Mila : L'IA agentique

Sprint Venture

Avantage IA : productivité dans la fonction publique

Jerry Huang

Publications

Communauté de pratique de Mila : L'IA agentique

Sprint Venture

Avantage IA : productivité dans la fonction publique

Mots-clés populaires:

Jerry Huang

Publications