paperlist.ai

AI conference papers

AI conference paper library

Search papers from 30+ top conferences, including NeurIPS, ICLR, ICML, CVPR, ACL and AAAI. Filter by keyword, then open any paper for its abstract, authors and links.

arXiv

DiaLLM: An Investigation into the Robustness-Generation Gap in English Dialect Adaptation

Jordan Painter, Dipankar Srirag, Adarsh Kappiyath, Diptesh Kanojia, Aditya Joshi, Lu Yin

Large language models increasingly \emph{understand} dialectal English, yet still \emph{produce} only standard, US-leaning English, leaving dialectal generation, the harder half of the problem, largely unaddressed. We introduce \textbf{DiaLLM}, which continually pretrains three open-weight language model families on the International Corpus of English and applies implicit and explicit post-training paradigms, each combined with three model alignment strategies, giving the first controlled comparison of these components across Australian, Indian, and Northern British English. Our results reveal that dialectal robustness and generation are \emph{dissociated}: benchmarks are shaped by continual pretraining and SFT, while alignment visibly reshapes generation in ways benchmarks do not capture. Explicit variety-targeted adaptation produces output reliably recognised as dialectal and preferred over broad alignment, yet the method that most aggressively optimises the dialectal reward is not preferred by human evaluators. Independent linguistic analysis corroborates this reward-quality gap, most clearly on two of the three families. No single alignment method dominates, and closing the gap will require richer reward designs and continued investment in dialectal resources. We release all code, checkpoints, and preference datasets.

arXiv

One Polluted Page Is Enough: Evaluating Web Content Pollution in LLM Recommenders

Minghao Luo, Liang Chen

Search-augmented LLMs increasingly mediate everyday consumer recommendations by retrieving live web content. This creates a new risk: LLM recommenders may consume web content that Generative Engine Optimization (GEO) operators have polluted to mislead them. We ask: to what extent do they become unwitting promoters of fake products? We introduce FORGE (Fake Online Recommendations in Generative Environments), which locally rewrites real products in a frozen set of retrieved web pages into fake ones and measures how often the LLM recommends the fake product, across 225 real products in 15 categories and 5 consumer scenarios. Across 12 commercial and open-weights LLMs, all models are vulnerable: a single polluted page yields fooled rates of up to 27%, while the full top-3 replacement raises this to 73.8%. Vulnerability varies across categories, increasing when models lack stable prior knowledge of the products. Reasoning does not mitigate this vulnerability; instead, it often generates spurious social proof to justify false recommendations. None of the four defenses is adequate: the skepticism prompt can exacerbate vulnerability much like reasoning, the two consensus filters risk suppressing legitimate products, and credibility re-ranking helps every model but removes only a sixth of the fakes. We release the FORGE benchmark and the evaluation code at https://github.com/leoluolol/forge-benchmark.

arXiv

CONSCIENTIA: Can LLM Agents Learn to Strategize? Emergent Deception and Trust in a Multi-Agent NYC Simulation

Aarush Sinha, Arion Das, Soumyadeep Nag, Charan Karnati, Shravani Nag, Chandra Vadhan Raj, Aman Chadha, Vinija Jain, Suranjana Trivedy, Amitava Das

As large language models (LLMs) are increasingly deployed as autonomous agents, understanding how strategic behavior emerges in multi-agent environments has become an important alignment challenge. We take a neutral empirical stance and construct a controlled environment in which strategic behavior can be directly observed and measured. We introduce a large-scale multi-agent simulation in a simplified model of New York City, where LLM-driven agents interact under opposing incentives. Blue agents aim to reach their destinations efficiently, while Red agents attempt to divert them toward billboard-heavy routes using persuasive language to maximize advertising revenue. Hidden identities make navigation socially mediated, forcing agents to decide when to trust or deceive. We study policy learning through an iterative simulation pipeline that updates agent policies across repeated interaction rounds using Kahneman-Tversky Optimization (KTO). Blue agents are optimized to reduce billboard exposure while preserving navigation efficiency, whereas Red agents adapt to exploit remaining weaknesses. Across iterations, the best Blue policy improves task success from 46.0% to 57.3%, although susceptibility remains high at 70.7%. Later policies exhibit stronger selective cooperation while preserving trajectory efficiency. However, a persistent safety-helpfulness trade-off remains: policies that better resist adversarial steering do not simultaneously maximize task completion. Overall, our results show that LLM agents can exhibit limited strategic behavior, including selective trust and deception, while remaining highly vulnerable to adversarial persuasion.

arXiv

Large Language Models Generate Harmful Responses Using a Distinct Mechanism, Shared Across Harm Types

Hadas Orgad, Boyi Wei, Kaden Zheng, Martin Wattenberg, Peter Henderson, Seraphina Goldfarb-Tarrant, Yonatan Belinkov

Large language models remain vulnerable to jailbreaks that elicit harmful responses, yet the mechanism behind harmful response generation is poorly understood. Here, we investigate how this capability is organized within model parameters. We identify and prune parameters that specifically support harmful compliance, providing a direct mechanistic analysis at the parameter level. We find that this capability depends on a sparse set of critical parameters: pruning these parameters substantially reduces harmful compliance while causing only limited degradation in benign capabilities, suggesting that key components of harmful generation are separable from those of general utility. Parameters identified from one harm category also reduce harmful responses in others, indicating components shared across harm types. This separability appears primarily in aligned models, suggesting that alignment training internally reshapes the harmful response mechanism even when behavioral safeguards remain brittle. We further show that harmful response generation is dissociable from the ability to recognize and reason about harmfulness. Finally, we extend our analysis to emergent misalignment and identify a sparse set of parameters contributing to it, with substantial sharing across fine-tuning domains. Together, these results reveal a consistent parameter-level organization underlying unsafe behaviors and point toward more principled interventions for improving model safety.

arXiv

PRAGMA: Revolut Foundation Model

Maxim Ostroukhov, Ruslan Mikhailov, Vladimir Iashin, Artem Sokolov, Andrei Akshonov, Vitaly Protasov, Andrey Goncharov, Dmitrii Beloborodov, Vince Mullin, Roman Yokunda Enzmann, Georgios Kolovos, Jason Renders, Pavel Nesterov, Anton Repushko

Modern financial systems generate vast quantities of transactional and event-level data that encode rich economic signals. This paper presents PRAGMA, a family of foundation models for banking event sequences. Our approach pre-trains a Transformer-based architecture with masked modelling on a large-scale, heterogeneous banking event corpus using a self-supervised objective tailored to the discrete, variable-length nature of financial records. The resulting model supports a wide range of downstream tasks such as credit scoring, fraud detection, and lifetime value prediction: strong performance can be achieved by training a simple linear model on top of the extracted embeddings and can be further improved with lightweight fine-tuning. Through extensive evaluation on downstream tasks, we demonstrate that PRAGMA achieves superior performance across multiple domains directly from raw event sequences, providing a general-purpose representation layer for financial applications.

arXiv

What Models Know, How Well They Know It: Knowledge-Weighted Fine-Tuning for Learning When to Say "I Don't Know"

Joosung Lee, Hwiyeol Jo, Donghyeon Ko, Kyubyung Chae, Cheonbok Park, Jeonghoon Kim

While large language models (LLMs) demonstrate strong capabilities across diverse user queries, they still suffer from hallucinations, often arising from knowledge misalignment between pre-training and fine-tuning. To address this misalignment, we reliably estimate a fine-grained, instance-level knowledge score via multi-sampled inference. Using the knowledge score, we scale the learning signal according to the model's existing knowledge, while encouraging explicit "I don't know" responses for out-of-scope queries. Experimental results show that this approach allows the model to explicitly express uncertainty when it lacks knowledge, while maintaining accuracy on questions it can answer. Furthermore, we propose evaluation metrics for uncertainty, showing that accurate discrimination between known and unknown instances consistently improves performance.

arXiv

DynHD: Hallucination Detection for Diffusion Large Language Models via Denoising Dynamics Deviation Learning

Yanyu Qian, Yue Tan, Yixin Liu, Wang Yu, Shirui Pan

Diffusion large language models (D-LLMs) have emerged as a promising alternative to auto-regressive models due to their iterative refinement capabilities. However, hallucinations remain a critical issue that hinders their reliability. To detect hallucination responses from model outputs, token-level uncertainty, such as entropy, has been widely used to indicate potential factual errors. Nevertheless, unlike auto-regressive models that generate tokens sequentially, D-LLMs generate fixed-length sequences simultaneously, where only a small subset of tokens is informative for hallucination detection. Thus, aggregating uncertainty over all tokens can be suboptimal. Moreover, the evolution trend of uncertainty throughout the diffusion process can also provide valuable signals, highlighting the necessity of modeling its denoising dynamics for hallucination detection. In this paper, we propose DynHD, which bridges these gaps from both spatial (token sequence) and temporal (denoising dynamics) perspectives. To handle the information density imbalance across tokens, we propose a semantic-aware evidence construction module that extracts hallucination-indicative signals by removing task-invariant structural tokens and emphasizing the uncertainty of the remaining informative tokens. To model denoising dynamics for hallucination detection, we introduce a reference evidence generator that learns the expected evolution trajectory of uncertainty evidence, along with a deviation-based hallucination detector that makes predictions by measuring the discrepancy between the observed and reference trajectories. Extensive experiments demonstrate that DynHD consistently outperforms state-of-the-art baselines while achieving higher efficiency across multiple benchmarks and backbone models. The code is available at: https://github.com/qyy11-com/DynHD.

arXiv

Safety Training May Persist Through Helpfulness Optimization in LLM Agents

Benjamin Plaut

Safety post-training has been studied extensively in single-step "chat" settings where safety typically refers to refusing harmful requests. We study an "agentic" (i.e., multi-step, tool-use) setting where safety refers to harmful actions directly taken by the LLM. We investigate the effects of using direct preference optimization (DPO) to optimize safety and/or helpfulness on the ToolEmu agentic benchmark. First, we find that safety training largely persists through subsequent helpfulness training. Second, we find a consistent negative linear correlation ($R^2 = 0.77$) between safety and helpfulness when considering all training configurations together. Even post-training on both metrics simultaneously simply results in another point on the same trend line rather than yielding a "best of both worlds" strategy, despite the presence of such strategies in our dataset. Overall, our findings underscore the need for a better understanding of post-training.

arXiv

Semantic Substrate Dynamics Theory: An Operator-Theoretic Framework for Geometric Semantic Drift

Stephen Russell

Studies of semantic drift report heterogeneous signals, including embedding displacement, neighbor change, distributional divergence, and recursive trajectory instability, without a shared account that relates them. Semantic Substrate Dynamics Theory (SSDT) treats these signals as observables of one time-indexed substrate, St = (X, dt, Pt), that couples embedding geometry to a local diffusion kernel. The contribution is commensurability with a mechanism layer: the substrate separates within-basin churn from basin crossing, recursion-induced instability, and intervention-order effects, distinctions that a single detection score does not recover. Coarse Ricci curvature functions as a dense structural descriptor of basin and bridge geometry across the graph, and bridge mass, a node-level aggregate of incident negative curvature, functions as a sparse descriptor of the genuine bridge structure that is typically uncommon in embedding graphs. For recursive generation, node displacement relative to an origin decomposes into a radial component and a tangential component, which separates bounded departure from continuing reinterpretation. The predictions are stated in falsifiable form with a pre-declared rejection rule, and the predicted leading indicator of future rewiring is a local density statistic rather than the curvature aggregate. This manuscript provides the formal model, the assumptions, the observable roles, and the test contracts; empirical performance is deferred.

arXiv

Dialects of Translationese Shape Language Model Learning

Jenny Kunz

Machine-translated data is widely used in multilingual NLP, particularly where native text is scarce. However, translated text differs systematically from native text. This phenomenon is known as translationese, and it reflects both traces of the source language and characteristic properties of translation itself. In this paper, we study how training on machine-translated data affects small English language models, focusing on how translationese from different source languages shapes linguistic acceptability judgments and language modeling for different domains. We train models on English text translated from 24 typologically and resource-diverse source languages, enabling a systematic analysis of how source language and corpus properties influence what models learn. Our results show that the source language has a clear impact on model behavior: general perplexity is more driven by the lexical diversity of the translated corpus, but grammatical performance is strongly correlated to typological similarity to English if trained on enough data. Even translation quality is a strong predictor of language modeling performance.

arXiv

MIRROR: A Multi-Agent Framework with Iterative Adaptive Revision and Hierarchical Retrieval for Optimization Modeling in Operations Research

Yifan Shi, Jiayi Wang, Minyi Wu, Ye Fan, Jialong Shi, Jianyong Sun

Operations Research (OR) relies on expert-driven modeling--a slow and fragile process ill-suited to novel scenarios. While large language models (LLMs) can automatically translate natural language into optimization models, existing approaches either rely on costly post-training or employ multi-agent frameworks, yet most still lack reliable collaborative error correction and task-specific retrieval, often leading to incorrect outputs. We propose MIRROR (a Multi-agent framework with Iterative adaptive Revision and hierarchical Retrieval for optimization modeling in Operations Research), a fine-tuning-free, end-to-end multi-agent framework that directly translates natural language optimization problems into mathematical models and solver code. MIRROR integrates two core mechanisms: (1) execution-driven iterative adaptive revision for automatic error correction, and (2) hierarchical retrieval to fetch relevant modeling and coding exemplars from a carefully curated exemplar library. Experiments show that MIRROR outperforms existing methods on standard OR benchmarks, with notable results on complex industrial datasets such as "IndustryOR" and "Mamo-ComplexLP". By combining precise external knowledge infusion with systematic error correction, MIRROR provides non-expert users with an efficient and reliable OR modeling solution, overcoming the fundamental limitations of general-purpose LLMs in expert optimization tasks.

arXiv

Lookahead-then-Verify: Reliable Constrained Decoding for Diffusion LLMs under Context-Free Grammars

Yitong Zhang, Yongmin Li, Yuetong Liu, Jia Li, Xiaoran Jia, Zherui Li, Ge Li

Diffusion Large Language Models (dLLMs) have demonstrated promising generative capabilities and are increasingly used to produce formal languages defined by context-free grammars, such as source code and chemical expressions. However, as probabilistic models, they still struggle to generate syntactically valid outputs reliably. A natural and promising direction to address this issue is to adapt constrained decoding techniques to enforce grammatical correctness during generation. However, applying these techniques faces two primary obstacles. On the one hand, the non-autoregressive nature of dLLMs renders most existing constrained decoding approaches inapplicable. On the other hand, current approaches specifically designed for dLLMs may allow intermediate outputs that are impossible to complete into valid sentences, which significantly limits their reliability in practice. To address these challenges, we present LAVE, a constrained decoding approach specifically designed for dLLMs. Our approach leverages a key property of dLLMs, namely their ability to predict token distributions for all positions in parallel during each forward pass. Whenever a new token is proposed by model, LAVE performs lookahead using these distributions to efficiently and reliably verify the validity of the proposed token. This design ensures reliable constraints by reliably preserving the potential for intermediate outputs to be extended into valid sentences. Extensive experiments across four widely used dLLMs and three representative benchmarks demonstrate that LAVE consistently outperforms existing baselines and achieves substantial improvements in syntactic correctness, while incurring negligible runtime overhead.

arXiv

FourierSampler: Unlocking Non-Autoregressive Potential in Diffusion Language Models via Frequency-Guided Generation

Siyang He, Qiqi Wang, Xiaoran Liu, Hongnan Ma, Yiwei Shi, Yuerong Song, Ying Zhu, Tianyi Liang, Zengfeng Huang, Ziwei He, Xipeng Qiu

Despite the non-autoregressive potential of diffusion language models (dLLMs), existing decoding strategies demonstrate positional bias, failing to fully unlock the potential of arbitrary generation. In this work, we delve into the inherent spectral characteristics of dLLMs and present the first frequency-domain analysis showing that low-frequency components in hidden states primarily encode global structural information and long-range dependencies, while high-frequency components are responsible for characterizing local details. Based on this observation, we propose FourierSampler, which leverages a frequency-domain sliding window mechanism to dynamically guide the model to achieve a "structure-to-detail" generation. FourierSampler outperforms other inference enhancement strategies on LLADA and SDAR, achieving relative improvements of 20.4% on LLaDA1.5-8B and 16.0% on LLaDA-8B-Instruct. It notably surpasses similarly sized autoregressive models like Llama3.1-8B-Instruct.

arXiv

An Empirical Study on Preference Tuning Generalization and Diversity Under Domain Shift

Constantinos Karouzos, Xingwei Tan, Nikolaos Aletras

Preference tuning aligns base language models to human judgments of quality, helpfulness, or safety by optimizing over explicit preference signals rather than likelihood alone. Prior work has shown that preference tuning degrades performance and reduces helpfulness outside the training domain. However, the extent to which adaptation strategies mitigate this domain shift remains unexplored. We address this challenge by conducting a comprehensive and systematic study of alignment generalization under domain shift. We compare five popular alignment objectives and various adaptation strategies from source to target, including target-domain supervised fine-tuning and pseudo-labeling, across summarization, question-answering helpfulness, and safety alignment tasks. Our findings reveal systematic differences in generalization across alignment objectives under domain shift. We show that adaptation strategies based on pseudo-labeling substantially reduce domain-shift degradation but induce mode collapse, revealing a generalization-diversity trade-off.

arXiv

Forgetting to Forget: Attention Sink as A Gateway for Backdooring LLM Unlearning

Bingqi Shang, Yiwei Chen, Yihua Zhang, Bingquan Shen, Sijia Liu

Large language model (LLM) unlearning is a key approach for removing undesired data, knowledge, or behaviors from pretrained models while retaining their general utility. Yet, with the rise of open-weight LLMs, we ask: can the unlearning process itself be backdoored, appearing successful under normal conditions yet reverting to pre-unlearned behavior when a hidden trigger is activated? Drawing inspiration from classical backdoor attacks that embed triggers into training data to enforce specific behaviors, we investigate backdooring unlearning, a setting in which models forget as intended in the clean setting but recover forgotten knowledge when the trigger appears. We show that designing such attacks presents unique challenges, hinging on where triggers are placed and how backdoor training is reinforced. We uncover a strong link between the backdoor efficacy and the attention sink phenomenon (i.e., shallow input tokens consistently attract disproportionate attention). Our analysis reveals that these attention sinks serve as gateways for backdooring unlearning: placing triggers at sink positions and aligning their attention values markedly enhances backdoor persistence. Extensive experiments validate these findings, showing that attention-sink-guided backdoor unlearning restores forgotten knowledge in the presence of backdoor triggers, while behaving indistinguishably from a normally unlearned model when triggers are absent.

arXiv

Beyond Benchmarks: LLM Evaluation with an Anthropomorphic and Lifecycle-oriented Roadmap

Jun Wang, Ninglun Gu, Kailai Zhang, Pengyong Li, Yelun Bao, Jin Yang, Xu Yin, Liwei Liu, Zijiao Zhang, Yihuan Liu, Gary G. Yen, Junchi Yan

Despite their rapid advancement, large language models (LLMs) suffer from a critical disconnect between benchmark scores and real-world utility. Current evaluation remains fragmented, prioritizing isolated technical metrics over the holistic, developmental, and societal aspects essential for deployment. Rather than serving merely as a descriptive catalog, this work establishes a diagnostic ontology that causally maps evaluation dimensions to the canonical LLM training pipeline, transforming evaluation from static ranking into a diagnostic tool for root-cause analysis. In this paper, we introduce an anthropomorphic evaluation framework that re-conceptualizes LLM capabilities through a four-dimensional lens: Intelligence Quotient (IQ), Professional Quotient (PQ), Emotional Quotient (EQ), and Value-oriented Quotient (VQ). We operationalize these concepts through a modular evaluation architecture and validate the framework's diagnostic claims through meta-analysis of public benchmark trends. Analyzing over 200 benchmarks, we synthesize key challenges and future directions. This work offers a strategic compass for developing LLMs that are not only technically proficient but also contextually relevant and ethically sound. A curated repository is available at: https://github.com/onejune2018/Awesome-LLM-Eval.

arXiv

From Isolation to Alignment: Unified LoRA for Efficient Multi-Task Learning

Jinda Liu, Yi Chang, Yuan Wu

Parameter-Efficient Fine-Tuning (PEFT) is essential for adapting Large Language Models (LLMs) to multi-task scenarios. A prevailing trend in this field involves complex LoRA variants with multiple adapters or heads, which rely on the premise that architectural isolation of task-specific knowledge is necessary. However, this design often introduces dynamic routing, preventing weight merging and causing significant inference latency. In this work, we present a direct challenge to this paradigm. We first reveal a paradox where a simplified, router-free multi-head model with high inter-head redundancy outperforms complex, diversity-driven baselines. Furthermore, we demonstrate that a unified, single-adapter LoRA with increased rank achieves highly competitive performance, questioning the necessity of multi-component structures. Based on these findings, we propose Align-LoRA, a unified and efficient framework that shifts the focus from architectural isolation to representation alignment. Align-LoRA incorporates an explicit alignment loss to encourage the learning of task-shared representations within a shared latent space. Crucially, our method maintains the standard LoRA architecture, ensuring zero inference latency via weight merging. Theoretical analysis and extensive experiments confirm that Align-LoRA significantly surpasses prevailing approaches, establishing a simpler, more effective, and production-friendly paradigm for multi-task PEFT. The code is available at https://github.com/jinda-liu/Align-LoRA.

arXiv

TELEVAL: A Benchmark Designed for Spoken Language Models in Chinese Interactive Scenarios

Zehan Li, Hongjie Chen, Qing Wang, Yuxin Zhang, Jing Zhou, Hang Lv, Mengjie Du, Yaodong Song, Jie Lian, Jian Kang, Jie Li, Yongxiang Li

Spoken Language Models (SLMs) are expected to support natural spoken interaction beyond task completion. However, existing SLM benchmarks primarily evaluate semantic correctness in structured settings and provide limited assessment of interactional behavior grounded in acoustic context. To address this gap, we introduce TELEVAL, a large-scale SLM benchmark for Chinese spoken interaction in instruction-free, audio-conditioned settings. TELEVAL evaluates two complementary aspects: (1) Reliable Content Fulfillment, which measures semantic accuracy of SLMs under diverse acoustic and linguistic conditions, and (2) Interactional Appropriateness, which assesses whether models produce natural and appropriate responses by implicitly grounding behavior in auditory cues. Experiments show that while models perform competitively on semantic tasks, their performance degrades under acoustic variability and in interactional settings. We observe consistent degradation from perceptual instability to interactional errors, and further identify a recurring failure pattern, termed the "Caption Trap", where models tend to describe perceived audio signals rather than produce appropriate interactive responses. These results indicate that current SLMs remain insufficiently aligned with the requirements of natural spoken interaction. TELEVAL provides a targeted framework for evaluating and analyzing interactional behavior in SLMs.

arXiv

Safety-Aligned Weights Are Not Enough: Refusal-Teacher-Guided Finetuning Enhances Safety and Downstream Performance under Harmful Finetuning Attacks

Seokil Ham, Yubin Choi, Yujin Yang, Seungju Cho, Younghun Kim, Changick Kim

While Finetuning-as-a-Service (FaaS) enables customization of Large Language Models (LLMs) using user data, this service is vulnerable to safety degradation when user data includes harmful prompts, a threat known as harmful finetuning attacks. To defend against this, prior work first constructs safety-aligned LLM and then finetunes the LLM on user data. However, we observe that the safety-aligned weights provide weak initialization for downstream task learning, leading to suboptimal safety and utility. Motivated by this limitation, we shift the safe FaaS finetuning paradigm from finetuning safety-aligned weights to finetuning base weights under explicit safety-teacher guidance. Specifically, we propose a Refusal-Teacher (Ref-Teacher)-guided finetuning framework. Our approach directly finetunes the base LLM under the guidance of a safety-aligned Ref-Teacher, which filters harmful prompts from user data and distills safety into the base LLM during finetuning. Extensive experiments demonstrate that our paradigm effectively reduces harmful outputs while improving finetuning performance on user-specific tasks.

arXiv

Scaling Electronic Health Record Foundation Models for Population Health Management

Liwen Sun, Hao-Ren Yao, Ophir Frieder, Xiang Qian, Chenyan Xiong

Population health management requires scalable methods to identify individuals at risk of chronic diseases such as cardiovascular conditions and cancer, yet existing approaches rely on fragmented data and resource-intensive screening. We present Scaling Electronic Health Record Foundation Models for Population Health Management, an Electronic Health Record Foundation Model that performs large-scale chronic disease prediction using cross-site longitudinal medical records. We pretrain Scaling Electronic Health Record Foundation Models for Population Health Management on billions of medical events from over 5 million patients across Taiwan and the United States, leveraging a unified code alignment framework to address cross-system heterogeneity, and characterize its scaling behavior via IsoFLOP analysis, training compute-optimal models up to 2.4B parameters. Across 11 chronic disease prediction tasks, Scaling Electronic Health Record Foundation Models for Population Health Management demonstrates strong scaling and generalization, outperforming tree-based models and both general and biomedical language models, achieving over 40% and 70% sensitivity at 99% specificity in U.S. and Taiwan cohorts, respectively. On the EHRShot benchmark, Scaling Electronic Health Record Foundation Models for Population Health Management surpasses prior EHR foundation models trained on in-site data despite substantial distribution shifts, highlighting strong few-shot generalization. Finally, we show that aligned cross-system data provides more effective pretraining signal than duplicating single-site data under data-limited settings, underscoring the importance of alignment for scalable healthcare modeling. Our analysis demonstrates the robustness of EHR-FM in various patient distributions and the benefits of operating in the ICD code space. The code will be open-sourced.

Keep browsing

More papers from top AI conferences

30+ top conferences, by conference, year, and topic.