Amortized Predictability-aware Training Framework for Time Series Forecasting and Classification
arXiv:2602.16224v1 Announce Type: new Abstract: Time series data are prone to noise in various domains, and training samples may contain low-predictability patterns that deviate from the normal data distribution, leading to training instability or convergence to poor local minima. Therefore,...
The academic article on the Amortized Predictability-aware Training Framework (APTF) does not directly address Immigration Law but may have indirect relevance to legal practice by offering insights into algorithmic bias mitigation and predictive accuracy improvements in data-driven decision-making. Specifically, the APTF’s design to identify and penalize low-predictability samples could inform legal analyses of algorithmic fairness in immigration-related systems (e.g., visa adjudication, risk assessment tools). The framework’s emphasis on mitigating model bias through amortization may signal broader trends in regulatory scrutiny of AI-driven administrative decisions, encouraging practitioners to anticipate scrutiny of predictive models’ reliability and equity in immigration contexts. Thus, while not immigration-specific, APTF contributes to the evolving discourse on algorithmic accountability that intersects with immigration law.
Title: Jurisdictional Comparison of Predictability-aware Training Frameworks in Immigration Law Practice The recent development of the Amortized Predictability-aware Training Framework (APTF) for time series forecasting and classification has sparked interest in the application of predictability-aware approaches in various fields, including immigration law practice. This commentary will compare the US, Korean, and international approaches to predictability-aware training frameworks, with a focus on their implications for immigration law practice. In the United States, the concept of predictability is closely tied to the principle of fairness in immigration law. The Supreme Court's decision in Plyler v. Doe (1982) established the principle that immigration laws must be applied in a manner that is free from arbitrary and capricious actions, promoting predictability and transparency in the decision-making process. In contrast, Korea's immigration law regime emphasizes the importance of predictability in the context of visa issuance and deportation proceedings. The Korean government has implemented various measures to improve the predictability of its immigration policies, including the introduction of a points-based system for visa applications. Internationally, the concept of predictability is increasingly recognized as a key principle in the administration of immigration laws. The United Nations High Commissioner for Refugees (UNHCR) has emphasized the importance of predictability in the context of refugee protection, highlighting the need for clear and transparent policies and procedures. Similarly, the European Union's Common Immigration Policy emphasizes the importance of predictability in the context of visa issuance and border management.
The article introduces a novel framework—APTF—addressing a critical gap in time series analysis by mitigating the impact of low-predictability samples. Practitioners in machine learning and time series modeling should consider integrating APTF’s Hierarchical Predictability-aware Loss (HPL) and amortization mechanisms to improve training stability and generalization, particularly in noisy datasets. While not directly tied to immigration law, parallels can be drawn to regulatory frameworks that penalize deviations from expected patterns (e.g., compliance monitoring), emphasizing the importance of adaptive, iterative evaluation in both domains. For deeper statutory or case law connections, consult legal experts on compliance or algorithmic bias issues.
What the Justice Department overlooks in its historical argument to end birthright citizenship
Immigration Matters is a recurring series by César Cuauhtémoc García Hernández that analyzes the court’s immigration docket, highlighting emerging legal questions about new policy and enforcement practices. In my last […]The postWhat the Justice Department overlooks in its historical argument...
The article is relevant to Immigration Law practice as it critically examines the Justice Department’s historical arguments against birthright citizenship, a foundational issue affecting citizenship eligibility and enforcement. Key developments include the scholarly critique of DOJ’s position on constitutional interpretation and its potential implications for litigation strategies defending birthright citizenship rights. The analysis signals heightened attention to constitutional debates in immigration enforcement, influencing advocacy and court arguments in related cases.
The recent debate on birthright citizenship in the United States, as highlighted in the article, has sparked a discussion on the jurisdictional approaches to this issue. In contrast to the US, which has a long-standing tradition of granting citizenship to children born on its territory (14th Amendment), Korea adheres to a more restrictive approach, only granting citizenship to children of Korean nationals or those born in Korea to foreign parents who have been lawfully resident in the country for at least five years. Internationally, the approach varies, with countries like Canada and the United Kingdom granting citizenship to children born on their territory, while others, such as Australia, have more restrictive laws. The implications of the US Justice Department's argument to end birthright citizenship, as highlighted in the article, would likely have significant consequences for immigration law practice in the country. It could lead to a reevaluation of the 14th Amendment and potentially alter the fundamental rights of children born in the US to non-citizen parents. This shift would be in contrast to the more restrictive approaches seen in countries like Korea, which may prioritize the rights of the parent or the child's connection to the country.
The article implicates statutory connections to the Fourteenth Amendment’s Citizenship Clause (Section 1) and regulatory implications under federal immigration enforcement frameworks, particularly as courts increasingly scrutinize administrative interpretations of citizenship. Practitioners should note that while birthright citizenship is entrenched in constitutional precedent (e.g., *United States v. Wong Kim Ark*, 1898), evolving DOJ arguments may influence litigation strategies in citizenship challenges, prompting heightened due diligence on constitutional and administrative law intersections. The series’ focus on emerging enforcement trends aligns with broader shifts in immigration jurisprudence, urging vigilance on precedent adaptation.
Call for Tutorial Proposals for CVPR 2026
This article does not appear to be relevant to Immigration Law practice area. However, I can comment on its irrelevance. The article is a call for proposals for tutorials at a computer vision and pattern recognition conference, CVPR 2026, and discusses the submission process, proposal requirements, and expectations for the tutorials. There is no mention of immigration law, policy, or regulations. In Immigration Law practice area, recent developments and policy changes may include: - The Biden Administration's efforts to reform the US immigration system, including proposals for a pathway to citizenship for certain undocumented immigrants. - The ongoing debate over the use of public charge rules in immigration decisions. - The impact of the COVID-19 pandemic on immigration policies and procedures, including the extension of certain immigration benefits and the implementation of new travel restrictions. This article does not provide any relevant information on these or other immigration law topics.
The article "Call for Tutorial Proposals for CVPR 2026" has no direct implications on Immigration Law practice, as it pertains to a computer vision conference. However, when comparing the approaches of the US, Korea, and international jurisdictions in the context of immigration law, several key differences emerge. In the US, immigration law is governed by the Immigration and Nationality Act (INA), which sets forth a complex framework for the admission and removal of non-citizens. In contrast, Korean immigration law is based on the Immigration Control Act, which provides for a more streamlined process for foreign nationals seeking to enter and reside in Korea. Internationally, the 1951 Refugee Convention and the 1990 Dublin Convention establish a framework for the treatment of refugees and asylum seekers, respectively. From an analytical perspective, the US approach to immigration law is often characterized by a more restrictive and complex framework, whereas Korea's approach is often seen as more welcoming and streamlined. Internationally, the 1951 Refugee Convention and the 1990 Dublin Convention reflect a more humanitarian approach to the treatment of refugees and asylum seekers. These jurisdictional differences have significant implications for the practice of immigration law, particularly in terms of the treatment of foreign nationals seeking to enter and reside in these jurisdictions.
As the Work Visa & Employment-Based Immigration Expert, I can provide domain-specific expert analysis of this article's implications for practitioners in the context of H-1B, L-1, O-1, and employment-based green cards. The article discusses a call for tutorial proposals for CVPR 2026, an event that may be relevant to foreign nationals working in the computer vision and pattern recognition field. Practitioners should note that the US Department of Labor's Occupational Information Network (O*NET) has designated Computer Vision Engineers and Pattern Recognition Specialists as high-demand occupations, potentially making them eligible for H-1B visas or other employment-based immigration benefits. In terms of case law, statutory, or regulatory connections, the article does not directly reference any specific immigration laws or regulations. However, the fact that a prominent conference like CVPR 2026 is seeking tutorial proposals may be relevant to practitioners who are planning to sponsor foreign nationals for H-1B or other employment-based immigration benefits. For example, if a US employer plans to sponsor a foreign national with expertise in computer vision and pattern recognition for an H-1B visa, they may need to demonstrate that the foreign national's work will be in a specialty occupation that is in short supply in the US labor market, which could be supported by evidence of the foreign national's participation in conferences like CVPR 2026. Practitioners should also be aware of the regulatory requirements for H-1B visas, including the requirement that
CVPR 2026 Call for Papers
This article appears to be unrelated to Immigration Law practice area. However, I can analyze it for relevance in other areas. Key legal developments, research findings, and policy signals in this article are not applicable to Immigration Law. The article is a call for papers for the Computer Vision and Pattern Recognition (CVPR) 2026 conference, focusing on computer vision and pattern recognition topics. If I were to stretch for relevance, I could say that the article touches on the intersection of technology and society, which might be of interest to immigration lawyers who deal with issues related to technology and global migration, such as the use of biometric data in border control or the impact of artificial intelligence on immigration decision-making. However, this is a very indirect and tenuous connection.
The CVPR 2026 Call for Papers, while focused on computer vision research, indirectly informs immigration law practice by influencing the development of technologies relevant to biometrics, surveillance, and data privacy—areas intersecting with immigration enforcement and border security. In the U.S., advancements in biometric identification may impact regulatory frameworks governing data collection, aligning with evolving privacy laws like the CPRA. South Korea’s stringent biometric data protections under the Personal Information Protection Act similarly shape compliance strategies for immigration-related tech. Internationally, the EU’s GDPR-driven approach underscores a global trend toward balancing innovation with individual rights, creating a shared imperative for legal practitioners to adapt to technological shifts affecting immigration law. Thus, even indirect research forums like CVPR contribute to shaping legal adaptation in immigration contexts.
As the Work Visa & Employment-Based Immigration Expert, I'll provide domain-specific expert analysis of the article's implications for practitioners in the context of H-1B, L-1, O-1, and employment-based green cards. The article's focus on computer vision and pattern recognition is closely related to the O-1 visa category, which requires evidence of extraordinary ability in the field. The list of topics of interest in computer vision and pattern recognition, such as deep learning architectures and techniques, image and video synthesis and generation, and multimodal learning, are areas where O-1 visa applicants may demonstrate their expertise and qualify for the visa. Practitioners should note that the O-1 visa category requires a labor certification or a job offer from a U.S. employer, and the beneficiary must demonstrate extraordinary ability in their field through evidence such as publications, awards, and recognition. The article's emphasis on original research and high-quality papers may be relevant to the O-1 visa process, as applicants may use their research and publications as evidence of their expertise. In terms of case law, statutory, or regulatory connections, the O-1 visa category is governed by 8 CFR 214.2(o), which outlines the requirements for the visa. The statute governing the O-1 visa is 8 U.S.C. 1101(a)(15)(O), which defines the category. The regulatory framework for the O-1 visa is also informed by the Immigration and Nationality Act
Statement Regarding API Security Incident | OpenReview
Based on the provided article, I found the following relevance to Immigration Law practice area: The article discusses a security incident involving unauthorized access to identities of reviewers, authors, and area chairs through a specific API endpoint. However, I couldn't find any direct relevance to Immigration Law practice area as the incident does not involve immigration-related data or policies. Nevertheless, the article may be relevant in a broader context of data protection and cybersecurity, which can be indirectly related to immigration law in cases where sensitive immigration data is involved. In the context of immigration law, this article may be seen as a signal for the importance of robust data protection measures to prevent unauthorized access to sensitive information. However, this is a very indirect connection, and the article primarily focuses on a security incident in the academic publishing domain. Key legal developments: The article highlights the importance of prompt action in responding to security incidents and the need for thorough analysis to understand the extent of the breach. Research findings: The article does not present any research findings but rather reports on a security incident and the actions taken to address it. Policy signals: The article does not signal any new policies but rather highlights the importance of data protection and cybersecurity in preventing unauthorized access to sensitive information.
### **Analytical Commentary: Data Security Incident in OpenReview’s API and Its Implications for Immigration Law Practice** The OpenReview API security breach—where unauthorized access exposed the identities of anonymous reviewers and authors—highlights critical **data protection and accountability gaps** in digital platforms, with direct implications for **immigration law practice** in the **US, South Korea, and under international frameworks**. While the US and South Korea both enforce strict data breach notification laws (e.g., **HIPAA and CCPA in the US; PIPA and PIPL in South Korea**), the **timely patching and forensic investigation** in this case reflect a **proactive, industry-led response** that contrasts with regulatory delays often seen in government systems. Internationally, under the **GDPR**, such a breach would trigger mandatory **72-hour reporting to authorities** and potential **fines up to 4% of global revenue**, whereas the US lacks a unified federal standard, relying instead on sector-specific laws. For immigration attorneys, this incident underscores the **vulnerability of biometric and identity data**—critical in visa processing—and the need for **enhanced cybersecurity due diligence** when handling sensitive client information, particularly in jurisdictions with weaker enforcement mechanisms. --- **Key Comparative Implications:** 1. **US Approach:** Relies on **sectoral laws** (e.g., HIPAA for health data, GLBA for financial data) and state
As the Work Visa & Employment-Based Immigration Expert, I'll provide domain-specific expert analysis of this article's implications for practitioners, focusing on potential connections to immigration law. The article discusses a security incident involving unauthorized access to sensitive information through a specific API endpoint. In the context of immigration law, this incident might be relevant to practitioners who deal with sensitive client information, particularly in cases involving H-1B, L-1, or O-1 petitions, where confidentiality and data protection are crucial. The security incident could be connected to the concept of "material misrepresentation" in immigration law, which is defined in 8 CFR 1001.19(d) as "any statement which is willfully made for the purpose of influencing the decision of the Service in the processing of an application or petition, which is false, or which contains any willfully false or misleading information concerning any material fact." Practitioners must ensure that they handle sensitive client information with utmost care to avoid any potential misrepresentation. Moreover, the incident highlights the importance of proper security measures to protect sensitive information, which is a key aspect of maintaining client confidentiality. This is particularly relevant in the context of employment-based immigration cases, where sensitive information about employees, employers, and clients is often involved. In terms of specific case law, the incident might be compared to the principles established in Matter of Hirschfeld (1988) 13 I&N Dec. 38, where the Board of Immigration Appeals (BIA) emphasized
AD-Bench: A Real-World, Trajectory-Aware Advertising Analytics Benchmark for LLM Agents
arXiv:2602.14257v1 Announce Type: new Abstract: While Large Language Model (LLM) agents have achieved remarkable progress in complex reasoning tasks, evaluating their performance in real-world environments has become a critical problem. Current benchmarks, however, are largely restricted to idealized simulations, failing...
The article *AD-Bench: A Real-World, Trajectory-Aware Advertising Analytics Benchmark for LLM Agents* is not directly relevant to **Immigration Law practice**, as it focuses on evaluating AI agents in advertising and marketing analytics rather than legal or policy frameworks. However, it may indirectly signal trends in **AI-driven legal tech** and **automated document analysis**, which could eventually intersect with immigration case management or regulatory compliance tools. For now, immigration practitioners should monitor AI advancements in adjacent fields but note that this study does not introduce legal or policy changes affecting immigration practice.
### **Jurisdictional Comparison & Analytical Commentary on AD-Bench’s Impact on Immigration Law Practice** The emergence of **AD-Bench**—a real-world benchmark for evaluating LLM agents in advertising analytics—has broader implications for **immigration law practice**, particularly in **automated legal decision-making, client intake, and document analysis**. While the U.S. and South Korea have taken divergent approaches to AI adoption in legal services, international frameworks (e.g., EU AI Act) provide a comparative lens. 1. **United States Approach**: The U.S. has adopted a **fragmented, case-by-case regulatory approach**, with agencies like USCIS and EOIR gradually integrating AI tools for visa processing and asylum adjudication. AD-Bench’s emphasis on **multi-tool collaboration** (L3 difficulty) mirrors U.S. immigration systems that require cross-referencing multiple databases (e.g., FBI, DHS, Interpol). However, unlike AD-Bench’s structured benchmarking, U.S. immigration AI adoption remains **ad hoc**, with concerns over **bias in algorithmic decision-making** (e.g., *Matter of A-B-*, 2018) leading to calls for stricter oversight. 2. **South Korean Approach**: South Korea’s immigration system, under the **Ministry of Justice (MOJ)**, has been more **centralized in AI adoption**, particularly in biometric screening and visa fraud detection.
As the Work Visa & Employment-Based Immigration Expert, I will analyze the article's implications for practitioners in the context of immigration law. The article discusses the development of a benchmark for evaluating Large Language Model (LLM) agents in complex advertising and marketing analytics tasks. While this may seem unrelated to immigration law, it has implications for practitioners who work with highly skilled foreign workers in specialized fields like data science and artificial intelligence. In the context of H-1B and L-1 visas, which are often used by employers to sponsor foreign workers in specialized fields, the AD-Bench benchmark may be relevant in demonstrating the qualifications and expertise of foreign workers. For instance, an employer may use AD-Bench as a tool to evaluate the abilities of a foreign worker in a data science or AI role, and demonstrate their qualifications to U.S. Citizenship and Immigration Services (USCIS) as part of an H-1B or L-1 petition. In terms of statutory or regulatory connections, this article may be relevant to the discussion around the "specialty occupation" definition in 8 CFR 214.2(h)(4)(ii), which defines a specialty occupation as one that requires theoretical and practical application of a body of highly specialized knowledge. The AD-Bench benchmark may be used to demonstrate that a foreign worker has the necessary expertise and qualifications to work in a specialty occupation, and therefore may be eligible for an H-1B or L-1 visa. Additionally, the article's discussion
Benchmark Leakage Trap: Can We Trust LLM-based Recommendation?
arXiv:2602.13626v1 Announce Type: new Abstract: The expanding integration of Large Language Models (LLMs) into recommender systems poses critical challenges to evaluation reliability. This paper identifies and investigates a previously overlooked issue: benchmark data leakage in LLM-based recommendation. This phenomenon occurs...
The academic article on LLM-based recommendation systems has indirect relevance to Immigration Law practice by highlighting systemic issues in evaluating algorithmic performance—specifically, data leakage in AI models can produce misleading metrics that affect decision-making. Key legal developments include the recognition that algorithmic bias or inaccuracy stemming from hidden data exposure may have implications for regulatory compliance, particularly in areas where AI is used for immigration eligibility assessments or recommendation platforms. The findings signal a growing need for transparency and validation protocols in AI-driven systems, prompting practitioners to consider potential legal risks associated with reliance on AI recommendations in client advising or administrative decision-making.
Jurisdictional Comparison and Analytical Commentary: The phenomenon of benchmark data leakage in Large Language Models (LLMs) poses significant implications for Immigration Law practice, particularly in the realm of asylum and refugee claims. In the US, for instance, the use of AI-powered tools to evaluate asylum claims could be compromised by data leakage, potentially leading to inaccurate determinations of refugee status. In contrast, Korea's approach to AI adoption in immigration decision-making is still in its nascent stages, and it remains to be seen how the country will address the issue of data leakage. Internationally, the International Organization for Migration (IOM) and other humanitarian organizations may need to reassess their reliance on AI-powered tools in refugee resettlement and protection efforts. In the context of Immigration Law, the impact of data leakage on LLM-based recommendation systems could be far-reaching. If LLMs are exposed to and memorize benchmark datasets, it could lead to artificially inflated performance metrics that fail to reflect true model performance. This could result in inaccurate determinations of refugee status, leading to potential human rights violations. Furthermore, the use of AI-powered tools in immigration decision-making raises concerns about transparency, accountability, and the potential for bias. In the US, the use of AI-powered tools in immigration decision-making is governed by the Administrative Procedure Act (APA) and the Immigration and Nationality Act (INA). However, the APA does not specifically address the issue of data leakage, and the INA does not provide clear guidelines for
The article on LLM-based recommendation data leakage raises critical implications for practitioners by highlighting a previously unrecognized vulnerability in evaluating AI performance. Specifically, the issue of data leakage—where LLMs are exposed to benchmark datasets during pre-training or fine-tuning—creates a misleading inflation of performance metrics, potentially distorting the perceived efficacy of models. From an immigration and legal perspective, practitioners advising on AI-related immigration petitions (e.g., O-1 for extraordinary ability or H-1B for specialty occupations) should be cognizant of the potential for inflated claims of AI capabilities due to such leakage, as it may affect the substantiation of expertise or technological innovation in petitions. Statutorily, this aligns with concerns under 8 U.S.C. § 1153(b)(2) regarding the requirement for genuine expertise, and case law such as Matter of Chawla may inform scrutiny of claims tied to AI performance metrics. Practitioners should integrate awareness of these evaluation pitfalls into due diligence for clients.
GREPO: A Benchmark for Graph Neural Networks on Repository-Level Bug Localization
arXiv:2602.13921v1 Announce Type: new Abstract: Repository-level bug localization-the task of identifying where code must be modified to fix a bug-is a critical software engineering challenge. Standard Large Language Modles (LLMs) are often unsuitable for this task due to context window...
The academic article on GREPO introduces a critical innovation for software engineering by establishing the first GNN benchmark for repository-level bug localization, addressing a persistent limitation in LLMs for large-scale code analysis. Key legal developments include the recognition of specialized algorithmic tools (like GNNs) over traditional retrieval methods in technical problem-solving, which may influence legal frameworks on intellectual property, software licensing, or algorithmic accountability. While not directly immigration-related, the research signals a broader policy trend toward validating specialized technical solutions as authoritative resources, potentially impacting regulatory approaches to AI governance or immigration-related tech workforce issues.
**Jurisdictional Comparison and Analytical Commentary on the Impact on Immigration Law Practice** The article "GREPO: A Benchmark for Graph Neural Networks on Repository-Level Bug Localization" has no direct implications on Immigration Law practice. However, a comparative analysis of US, Korean, and international approaches to innovation and technology adoption in the context of Immigration Law can be insightful. In the US, the H-1B visa program allows foreign workers with specialized skills, including software engineers, to work in the country. The US government has been exploring ways to streamline the H-1B application process, including the use of artificial intelligence (AI) and machine learning (ML) to improve efficiency and accuracy. The development of GREPO, a benchmark for Graph Neural Networks (GNNs) on repository-level bug localization tasks, could potentially be applied to improve the H-1B application process by automating tasks and reducing processing times. In Korea, the government has implemented various initiatives to promote innovation and technology adoption in the country. The Korean Immigration Service has introduced an online application system for visa applications, which utilizes AI and ML to streamline the process. The development of GREPO could be applied to improve the Korean immigration system by enhancing the accuracy and efficiency of visa application processing. Internationally, the United Nations High Commissioner for Refugees (UNHCR) has been exploring the use of AI and ML to improve the refugee resettlement process. The development of GREPO could potentially be applied to improve the UNHCR
The article introduces GREPO as a pivotal benchmark for GNNs in repository-level bug localization, addressing a critical gap in software engineering research. By providing a scalable dataset (86 Python repositories, 47294 bug-fixing tasks) tailored for GNN processing, GREPO enables direct application of graph-based models, potentially shifting the paradigm from traditional retrieval methods (e.g., keyword matching, text similarity) to more sophisticated GNN-driven solutions. Practitioners in software engineering and AI/ML should note this as a foundational resource; its impact aligns with regulatory trends promoting innovation in AI-driven software maintenance (e.g., USPTO’s focus on AI applications in engineering). Case law relevance may emerge if GREPO’s methodology influences patent eligibility for AI-assisted bug detection under 35 U.S.C. § 101.
SCOTUStoday: Sotomayor criticizes Kavanaugh
Curious about how Supreme Court justices spend their spare time? Justice Sonia Sotomayor revealed on Tuesday that she likes reading … recent books from her colleagues. She “said she just […]The postSCOTUStoday: Sotomayor criticizes Kavanaughappeared first onSCOTUSblog.
Multi-objective Evolutionary Merging Enables Efficient Reasoning Models
arXiv:2604.06465v1 Announce Type: new Abstract: Reasoning models have demonstrated remarkable capabilities in solving complex problems by leveraging long chains of thought. However, this more deliberate reasoning comes with substantial computational overhead at inference time. The Long-to-Short (L2S) reasoning problem seeks...
State-of-the-Art Arabic Language Modeling with Sparse MoE Fine-Tuning and Chain-of-Thought Distillation
arXiv:2604.06421v1 Announce Type: new Abstract: This paper introduces Arabic-DeepSeek-R1, an application-driven open-source Arabic LLM that leverages a sparse MoE backbone to address the digital equity gap for under-represented languages, and establishes a new SOTA across the entire Open Arabic LLM...
State election dispute on political speech comes to Supreme Court on interim docket
Lawyers for Ohio Secretary of State Frank LaRose, as well as county election officials, urged the Supreme Court on Wednesday to let them go ahead with a ballot that does […]The postState election dispute on political speech comes to Supreme...
A Benchmark of Classical and Deep Learning Models for Agricultural Commodity Price Forecasting on A Novel Bangladeshi Market Price Dataset
arXiv:2604.06227v1 Announce Type: new Abstract: Accurate short-term forecasting of agricultural commodity prices is critical for food security planning and smallholder income stabilisation in developing economies, yet machine-learning-ready datasets for this purpose remain scarce in South Asia. This paper makes two...
Application-Driven Pedagogical Knowledge Optimization of Open-Source LLMs via Reinforcement Learning and Supervised Fine-Tuning
arXiv:2604.06385v1 Announce Type: new Abstract: We present an innovative multi-stage optimization strategy combining reinforcement learning (RL) and supervised fine-tuning (SFT) to enhance the pedagogical knowledge of large language models (LLMs), as illustrated by EduQwen 32B-RL1, EduQwen 32B-SFT, and an optional...
Blending Human and LLM Expertise to Detect Hallucinations and Omissions in Mental Health Chatbot Responses
arXiv:2604.06216v1 Announce Type: new Abstract: As LLM-powered chatbots are increasingly deployed in mental health services, detecting hallucinations and omissions has become critical for user safety. However, state-of-the-art LLM-as-a-judge methods often fail in high-risk healthcare contexts, where subtle errors can have...
The Illusion of Superposition? A Principled Analysis of Latent Thinking in Language Models
arXiv:2604.06374v1 Announce Type: new Abstract: Latent reasoning via continuous chain-of-thoughts (Latent CoT) has emerged as a promising alternative to discrete CoT reasoning. Operating in continuous space increases expressivity and has been hypothesized to enable superposition: the ability to maintain multiple...
AgentOpt v0.1 Technical Report: Client-Side Optimization for LLM-Based Agent
arXiv:2604.06296v1 Announce Type: new Abstract: AI agents are increasingly deployed in real-world applications, including systems such as Manus, OpenClaw, and coding agents. Existing research has primarily focused on \emph{server-side} efficiency, proposing methods such as caching, speculative execution, traffic scheduling, and...
SHAPE: Stage-aware Hierarchical Advantage via Potential Estimation for LLM Reasoning
arXiv:2604.06636v1 Announce Type: new Abstract: Process supervision has emerged as a promising approach for enhancing LLM reasoning, yet existing methods fail to distinguish meaningful progress from mere verbosity, leading to limited reasoning capabilities and unresolved token inefficiency. To address this,...
A Parameter-Efficient Transfer Learning Approach through Multitask Prompt Distillation and Decomposition for Clinical NLP
arXiv:2604.06650v1 Announce Type: new Abstract: Existing prompt-based fine-tuning methods typically learn task-specific prompts independently, imposing significant computing and storage overhead at scale when deploying multiple clinical natural language processing (NLP) systems. We present a multitask prompt distillation and decomposition framework...
Spectral Edge Dynamics Reveal Functional Modes of Learning
arXiv:2604.06256v1 Announce Type: new Abstract: Training dynamics during grokking concentrate along a small number of dominant update directions -- the spectral edge -- which reliably distinguishes grokking from non-grokking regimes. We show that standard mechanistic interpretability tools (head attribution, activation...
Beyond Facts: Benchmarking Distributional Reading Comprehension in Large Language Models
arXiv:2604.06201v1 Announce Type: new Abstract: While most reading comprehension benchmarks for LLMs focus on factual information that can be answered by localizing specific textual evidence, many real-world tasks require understanding distributional information, such as population-level trends and preferences expressed across...
In-Context Learning in Speech Language Models: Analyzing the Role of Acoustic Features, Linguistic Structure, and Induction Heads
arXiv:2604.06356v1 Announce Type: new Abstract: In-Context Learning (ICL) has been extensively studied in text-only Language Models, but remains largely unexplored in the speech domain. Here, we investigate how linguistic and acoustic features affect ICL in Speech Language Models. We focus...
ART: Attention Replacement Technique to Improve Factuality in LLMs
arXiv:2604.06393v1 Announce Type: new Abstract: Hallucination in large language models (LLMs) continues to be a significant issue, particularly in tasks like question answering, where models often generate plausible yet incorrect or irrelevant information. Although various methods have been proposed to...
STDec: Spatio-Temporal Stability Guided Decoding for dLLMs
arXiv:2604.06330v1 Announce Type: new Abstract: Diffusion Large Language Models (dLLMs) have achieved rapid progress, viewed as a promising alternative to the autoregressive paradigm. However, most dLLM decoders still adopt a global confidence threshold, and do not explicitly model local context...
Databricks co-founder wins prestigious ACM award, says ‘AGI is here already’
Matei Zaharia has won the top honor from the Association for Computing Machinery. Now he's working on AI for research and says AGI is simply misunderstood.
A Severity-Based Curriculum Learning Strategy for Arabic Medical Text Generation
arXiv:2604.06365v1 Announce Type: new Abstract: Arabic medical text generation is increasingly needed to help users interpret symptoms and access general health guidance in their native language. Nevertheless, many existing methods assume uniform importance across training samples, overlooking differences in clinical...
Does a Global Perspective Help Prune Sparse MoEs Elegantly?
arXiv:2604.06542v1 Announce Type: new Abstract: Empirical scaling laws for language models have encouraged the development of ever-larger LLMs, despite their growing computational and memory costs. Sparse Mixture-of-Experts (MoEs) offer a promising alternative by activating only a subset of experts per...
ODE-free Neural Flow Matching for One-Step Generative Modeling
arXiv:2604.06413v1 Announce Type: new Abstract: Diffusion and flow matching models generate samples by learning time-dependent vector fields whose integration transports noise to data, requiring tens to hundreds of network evaluations at inference. We instead learn the transport map directly. We...
Distributed Interpretability and Control for Large Language Models
arXiv:2604.06483v1 Announce Type: new Abstract: Large language models that require multiple GPU cards to host are usually the most capable models. It is necessary to understand and steer these models, but the current technologies do not support the interpretability and...
MO-RiskVAE: A Multi-Omics Variational Autoencoder for Survival Risk Modeling in Multiple MyelomaMO-RiskVAE
arXiv:2604.06267v1 Announce Type: new Abstract: Multimodal variational autoencoders (VAEs) have emerged as a powerful framework for survival risk modeling in multiple myeloma by integrating heterogeneous omics and clinical data. However, when trained under survival supervision, standard latent regularization strategies often...