AI Research & Papers — September 28, 2026

Findings

1. [1] 2026-09-28 One year in: How Microsoft Research Asia – Singapore is advancing research, partnership and talent for real-world impact

   On July 24, 2025, Microsoft Research Asia – Singapore (MSRA – Singapore) opened its doors as Microsoft’s first research lab in Southeast Asia. The launch built on more than two decades of collaboration between Microsoft Research Asia and Singapore’s universities,… Government engagement extends beyond EDB. In July 2025, IMDA (opens in new tab) co-hosted Microsoft Research Asia Singapore Day (opens in new tab) — Advancing Industrial AI, a full-day event spanning healthcare AI, interactive world simulators, AI agents, and societal AI.… The IPP represents a sustained form of collaboration, enabling PhD students to conduct long-term research under joint mentorship from academia and industry. Qiming Huang, the first IPP student at MSRA — Singapore, is co-supervised by NUS professor Mike Shou and MSRA…

2. [2] 2026-09-28 New formulation helps RNA vaccines withstand high temperatures

   RNA vaccines, which have been proven effective against Covid-19, are now being developed for many other diseases, including cancer. One of the drawbacks to these vaccines is that they require ultracold storage, but researchers from MIT have found a promising… Mice that were vaccinated with these particles, even after long-term storage, showed equivalent immune responses to mice that received vaccines carried by LNPs similar to the original Moderna formulation.The researchers also used their new heat-resistant formulation to create solid microneedle…

3. [3] 2026-09-28 All In Good Time: Causality-Aware Framework for LLM-Based Simultaneous Speech-to-Speech Translation

   arXiv:2609.30416v1 Announce Type: new Abstract: Large Language Models (LLMs) have shown strong performance in low-resource offline translation; however, extending them to simultaneous speech-to-speech translation (Simul-S2ST) remains challenging due to the scarcity of causally aligned training data with high cross-lingual speaker fidelity. In addition, existing approaches rely on fixed translation policy or confidence heuristics, leading to suboptimal quality and higher latency.…

4. [4] 2026-09-28 Same Text, Different Numbers: The Divergence of LLM-Based Measures

   arXiv:2609.31013v1 Announce Type: cross Abstract: Researchers increasingly use generative large language models (LLMs) to convert corporate text into empirical variables. We examine the extent to which LLM-based textual measures are invariant to model choice using thirteen measures, including sentiment, management clarity, uncertainty, answer specificity, and climate and political risk. Seven LLMs from different providers score earnings call transcripts of S&P…

5. [5] 2026-09-28 A Unified Account of Concepts and Chunks

   arXiv:2609.30414v1 Announce Type: new Abstract: Cognitive psychology has studied how people encode, use, and learn concepts that describe categories, and how they represent, recognize, and acquire chunks for familiar patterns of elements. The literatures on these two topics are nearly disjoint, which poses a challenge for unified theories of cognition. In this paper, we review Cobweb, a computational account of…

6. [6] 2026-09-28 Thinking Less to Simulate Better: Intuitive Prompting Improves LLM Agents Simulating Individual Social Media Reactions, Including Unfamiliar Content

   arXiv:2609.30563v1 Announce Type: cross Abstract: Platform policies are increasingly tested on artificial users, making agent fidelity important. Yet convincing fake profiles could also manipulate perceived public opinion before elections. Validation has concentrated on agreement with human behaviour and has paid little attention to whether an agent behaves in line with the profile it was given. The present study profiled eight…

7. [7] 2026-09-28 The Communication Map of a Transformer

   arXiv:2608.22007v2 Announce Type: replace-cross Abstract: The components of a transformer communicate by writing to and reading from a shared residual stream, and the mechanistic interpretability literature has mapped these connections by hand, one circuit at a time. We present the communication map, which charts every potential communication channel from the geometry of the model's weights alone, generalizing the composition score…

8. [8] 2026-09-28 A Benchmark Framework for Screening Automation in Systematic Reviews

   arXiv:2609.30298v1 Announce Type: new Abstract: Systematic reviews (SR) are essential for evidence-based research, but their screening phase is highly time-consuming and labor-intensive. Large language models (LLMs) offer a promising opportunity to reduce this workload by assisting with article relevance classification. However, existing evaluation approaches often rely on traditional metrics that may be misleading for highly imbalanced SR screening datasets.This paper…

9. [9] 2026-09-28 Not All Memories Are Equal: Hierarchical Collaborative Memory for Validity-Aware Retrieval in LLM Agents

   arXiv:2609.30289v1 Announce Type: new Abstract: In team collaboration scenarios, memory is heterogeneous and continually evolving. Team memories capture collective decisions, protocols, and current consensus, while individual memories preserve member-specific observations, execution traces, and intermediate progress. Existing memory-augmented systems typically retrieve from all stored memories as a flat pool, ranking them by semantic relevance, importance, or recency without modeling hierarchical structure…

10. [10] 2026-09-28 Auditing and Repairing LLM-as-Judge Failures in a Production Text-to-SQL Pipeline

    arXiv:2609.30290v1 Announce Type: new Abstract: Production text-to-SQL pipelines often end with an LLM-as-judge whose agreement with human annotators has never actually been measured. When we checked ours, the deployed gpt-4o-mini judge agreed with two-author gold at only Cohen's kappa = 0.04 on a disagreement-enriched set and 0.42 on a uniform-random spot-check, over-flagging 77.1% of the human-FAITHFUL cases in the enriched…

11. [11] 2026-09-28 A Survey on Fake Review Detection: From Pre-trained Language Models to Large Language Models

    arXiv:2609.30292v1 Announce Type: new Abstract: Online reviews shape consumer decisions, platform governance, and corporate reputation.Fake reviews compromise this information channel by injecting deceptive evidence into rating systems, recommendation pipelines, and public trust mechanisms.The rise of large language models, or LLMs, has changed the problem in two directions.LLMs can generate fluent and context-aware deceptive reviews, while pre-trained language models, or PLMs,…

12. [12] 2026-09-28 SlideLab: Audience-Centered Scientific Slide Generation and Evaluation

    arXiv:2609.30294v1 Announce Type: new Abstract: Scientific presentations are more than summaries of research papers. They need to present the work in a coherent sequence, explain the main ideas clearly, and help the audience follow the presentation. We present SlideLab, a training-free multi-agent framework for generating scientific presentations from research papers. SlideLab first plans the presentation narrative, then builds and iteratively…

13. [13] 2026-09-28 Cartograph: Federated Tool Discovery with Operator-Attested Retrieval for AI Agents

    arXiv:2609.30293v1 Announce Type: new Abstract: The Model Context Protocol (MCP) enables AI agents to discover and call tools, but loading every definition becomes expensive as connected catalogs grow. We present Cartograph, a federated MCP proxy that changes agent-visible tool discovery from $O(n)$ catalog traversal to $O(k)$ progressive disclosure. Cartograph combines three mechanisms: (1) operator-attested capability cards, Ed25519-signed descriptions generated under…

14. [14] 2026-09-28 Bootstrapping Conversational Recommendation Agents At Spotify: Synthetic Data Generation and Self-Improvement Loops

    arXiv:2609.30297v1 Announce Type: new Abstract: Conversational recommendation agents are a new paradigm for content discovery, enabling users to express complex intents through natural language (e.g., "recommend Italian indie artists I haven't heard before"). A central challenge in building such agents is optimizing agent planning — deciding how to select, sequence, and invoke tools — particularly in cold-start settings where real…

15. [15] 2026-09-28 SignTrace: Describe a Sign, Find the Word

    arXiv:2609.30295v1 Announce Type: new Abstract: Identifying an unfamiliar sign is difficult when a learner remembers its movement but does not know its meaning or formal feature codes. SignTrace addresses this longstanding reverse-lookup problem through natural-language access to a Chinese sign-language dictionary. The system integrates LLM-based dictionary enrichment, action extraction, dictionary-style rewriting, seven-channel retrieval, and candidate reranking over 6,699 entries. It…

16. [16] 2026-09-28 The Right Information Extraction Pipeline Depends on the Document: Accuracy-Energy Trade-offs for Small, Local Models

    arXiv:2609.31341v1 Announce Type: cross Abstract: Whether an information extraction pipeline should process page images or parsed text depends on the document, and the answer flips across the layout spectrum. We study this trade-off under a constraint that rules out (closed) cloud services: privacy-sensitive documents processed on-premise by small ($le 8mathrm{B}$ parameter) text-only and vision–language models, evaluated on both accuracy and…

17. [17] 2026-09-28 Symbiotic Architecture for Post-Hoc Audio Extension of Frozen Language Models

    arXiv:2609.30784v1 Announce Type: cross Abstract: This paper proposes an architecture for equipping large language models (LLMs) with audio-understanding capabilities without fine-tuning their weights. The proposed symbiotic architecture employs an injector module that writes audio-conditioned vectors directly into the target LLM's short-term memory, i.e., the key-value (KV) cache, enabling the LLM to behave as an audio language model (ALM). The architectural…

18. [18] 2026-09-28 VLAA-GUI: Knowing When to Stop, Recover, and Search, A Modular Framework for GUI Automation

    arXiv:2604.21375v3 Announce Type: replace Abstract: Autonomous GUI agents face two fundamental challenges: early stopping, where agents prematurely declare success without verifiable evidence, and repetitive loops, where agents cycle through the same failing actions without recovery. We present VLAA-GUI, a modular GUI agentic framework built around three integrated components that guide the system on when to Stop, Recover, and Search. First,…

19. [19] 2026-09-28 Learning to Stop without Learning to Stop: Self-Supervised Confidence Training Improves Reasoning Efficiency

    arXiv:2609.31619v1 Announce Type: cross Abstract: Reasoning models often generate very long reasoning traces, making inference computationally expensive. Existing approaches typically improve efficiency either through inference-time early-stopping mechanisms or by explicitly encouraging shorter reasoning during training, for example through reinforcement learning with length penalties. We show that substantial efficiency gains can instead emerge from a different kind of supervision: textit{confidence}. Using…

20. [20] 2026-09-28 ArGuard Shared Task: Harmful Content Detection in Arabic Memes and LLM Prompts

    arXiv:2609.29349v2 Announce Type: replace Abstract: ArGuard is a shared task on harmful content detection in Arabic memes and LLM prompts. It includes two tracks: Track A focuses on multimodal hate detection in Arabic memes, while Track B addresses harmful prompt detection for Arabic LLM safety evaluation. In total, 58 teams registered, 35 participated in the final evaluation, and 27 submitted…

21. [21] 2026-09-28 Attention-Discounted Adaptive Sampler for Masked Diffusion Language Models

    arXiv:2606.10829v5 Announce Type: replace Abstract: Masked diffusion language models can reduce inference steps by revealing multiple tokens per denoising iteration, but this parallelism is fragile: positions that are individually confident may be unsafe to commit together when their predictions are coupled. Existing training-free samplers such as Top-(k), Fast-dLLM, and EB-Sampler mainly control how many tokens to reveal, while often ranking…

22. [22] 2026-09-28 FlyAOC: Evaluating Agentic Ontology Curation of Drosophila Scientific Knowledge Bases

    arXiv:2602.09163v2 Announce Type: replace-cross Abstract: Scientific knowledge bases accelerate discovery by curating findings from primary literature into structured, queryable formats for both human researchers and emerging AI systems. Maintaining these resources requires expert curators to search papers, reconcile evidence across documents, and produce ontology-grounded annotations. Existing benchmarks usually evaluate isolated subtasks, such as named entity recognition or relation extraction, and…

23. [23] 2026-09-28 Rethinking Data Quality for AI-Driven Systems: Evidence from Practitioner Interviews

    arXiv:2609.31191v1 Announce Type: cross Abstract: Data quality research has usually treated data as an input that is stored, processed, and validated. In AI-driven software-intensive systems, data also shapes model behavior, evaluation, and lawful use. Empirical evidence remains limited on how practitioners define, assess, and manage quality under these conditions. We interviewed 16 practitioners from nine organizations and analyzed the transcripts…

24. [24] 2026-09-28 BioEVAL: A global, multi-institutional benchmark of large language and multimodal models for bioengineering

    arXiv:2609.30489v1 Announce Type: new Abstract: Large Language Models (LLMs) have demonstrated historic breakthroughs in general reasoning with early successes in biomedical science. However, existing LLM benchmarking emphasizes factual recall, offering limited insight into model performance on frontier and multimodal tasks. We assembled BioEVAL (BioEngineering Validation of AI and LLMs), a global, multi-institutional initiative designed to assess experimental reasoning capability across…

25. [25] 2026-09-28 PolyChirp: Multi-Species Birdsong Classification Using TinyML on Low-Power Acoustic Sensors

    arXiv:2608.23101v2 Announce Type: replace-cross Abstract: Recent progress in the field of TinyML has demonstrated that low-power hardware based on microcontrollers can achieve bird species monitoring in real time based on acoustic sensor data for an entire breeding period on a single battery charge. However, the state of the art on low-power microcontrollers was so far limited to binary classification of…

26. [26] 2026-09-28 Do LLMs Understand Context? A Knowledge Graph-Based Evaluation Framework

    arXiv:2609.30484v1 Announce Type: new Abstract: While large language models (LLMs) have achieved remarkable linguistic capabilities, a profound question lingers at their core: do these models truly comprehend context or simply excel at pattern matching on an unprecedented scale? Contextual understanding in LLMs refers to the ability to correctly extract relevant information from a given context, integrate it into a coherent…

27. [27] 2026-09-28 An AI Agent Execution Environment to Safeguard User Data

    arXiv:2604.19657v3 Announce Type: replace-cross Abstract: AI agents promise to serve as general-purpose personal assistants for their users, which requires them to have access to private user data (e.g., personal and financial information). This poses a serious risk to security and privacy: an AI model may hallucinate or make mistakes, and adversaries may attack it (e.g., via prompt injection) to exfiltrate…

28. [28] 2026-09-28 Pretrained ASR Pseudo-labeling for Noisy Police Audio

    arXiv:2609.30469v1 Announce Type: new Abstract: Pretrained ASR systems perform poorly on noisy Broadcast Police Communication (BPC), hindering efforts to understand police decision-making. Pseudo-labeling offers an unsupervised path to improve ASR without expensive human labels, but the efficacy of this approach on very noisy domains is not known. In this work, we systematically assess the opportunities and limits of pseudo-labeling to…

29. [29] 2026-09-28 Consist-Retinex: One-Step Noise-Emphasized Consistency Training Accelerates High-Quality Retinex Enhancement

    arXiv:2512.08982v4 Announce Type: replace-cross Abstract: Retinex-based low-light image enhancement benefits from separating reflectance and illumination, yet recent generative approaches often rely on iterative sampling and are difficult to deploy under strict latency budgets. Consistency models offer a natural route to one-step restoration, but direct adaptation to Retinex-factorized enhancement is unstable: one-step inference is evaluated at the high-noise endpoint, whereas standard…

30. [30] 2026-09-28 When Is a Multi-Agent Code Judge Actually Grounded? Two Label-Free Measurements, and a Judge That Declines to Guess

    arXiv:2609.30328v1 Announce Type: new Abstract: When one language model judges whether another's code is correct, it does not report the absence of evidence. It returns a confident verdict with reasoning attached, indistinguishable from a verdict it had grounds for. Multi-agent verification, which decomposes a judgment into checkable claims and verifies each against evidence, is a promising response and works well…

31. [31] 2026-09-28 Bridging LLM Agents and Data Spaces: An Architectural Mediation Approach using the Model Context Protocol

    arXiv:2609.30341v1 Announce Type: new Abstract: Data Spaces enable sovereign and governed data sharing across organizational boundaries, but their integration with AI agents remains challenging due to mismatches between probabilistic language model interactions and policy-driven data infrastructures. This article presents an architectural mediation approach based on the Model Context Protocol (MCP), implemented through the Eunomia Agent, to enable controlled interaction between…

32. [32] 2026-09-28 Stealth Apart, Harm Together: Skill Cascading Attacks on Skill-Based Agent Systems

    arXiv:2609.30383v1 Announce Type: new Abstract: A skill is a modular package of natural-language instructions, executable scripts, and reference resources that an agent can load at runtime to extend its capabilities for a specific task. Skill-based agent systems therefore enable flexible reuse of third-party capabilities, but the openness of this skill ecosystem also opens up a new attack surface. Prior work…

33. [33] 2026-09-28 Predicting Transmembrane Protein Topology from 3D Structure

    arXiv:2609.30446v1 Announce Type: new Abstract: This paper presents a novel approach to infer protein topology using the state-of-the-art graph neural network (GNN), SchNet. The model is trained on the same dataset used to develop the recent DeepTMHMM model with 5-fold cross-validation. Unlike the conventional approaches based on using only the protein sequences or the $alpha$-carbons as features, we have decoded…

34. [34] 2026-09-28 A Synthetic Ground-Truth Framework for the Evaluation of Explainable AI Methods

    arXiv:2609.30397v1 Announce Type: new Abstract: Evaluating explainable Artificial Intelligence (XAI) methods is a challenging task due to the lack of reliable evaluation procedures and, in particular, the absence of ground truth explanations. In the literature, existing evaluation approaches typically assess explanations by measuring their fidelity with respect to the predictions of a black-box model. However, such evaluation strategies only quantify…

35. [35] 2026-09-28 Policy Regret for Embedding Model Routing: Contextual Bandits with Low-Rank Experts

    arXiv:2606.14929v2 Announce Type: replace-cross Abstract: Modern recommendation systems increasingly rely on dynamically routing diverse queries to multiple embedding models. Despite its practical significance, this problem remains poorly understood under realistic conditions like adversarial queries, bandit feedback, and limited observability of models. We formalize embedding model routing as an adversarial contextual linear bandit with low-rank experts, where contexts are queries, actions…

36. [36] 2026-09-28 Spectral Feedback for Test-Time Alignment of Protein Diffusion Models

    arXiv:2609.30456v1 Announce Type: new Abstract: Reward maximization alignment methods for discrete diffusion models have primarily focused on steering the reverse process, either by influencing token logits or by selecting favorable sequences at intermediate steps. These approaches largely treat inference as a unidirectional process, lacking mechanisms for revisiting undesirable token selections. We introduce Spectral Feedback, an algorithm that selects edit-positions in…

37. [37] 2026-09-28 UniAR: A Unified Framework for Autism Recognition Enhanced by Multi-View Prompt Learning

    arXiv:2609.31298v1 Announce Type: cross Abstract: Autism Spectrum Disorder (ASD) is a complex neurodevelopmental disorder for which early and accurate diagnosis is critical to improving long-term developmental outcomes. However, existing ASD recognition methods are often constrained by the scarcity of diagnostic text data, forcing them to rely mainly on visual analysis and limiting their ability to model clinically meaningful semantic reasoning.…

38. [38] 2026-09-28 Lifted Bellman Linear Programming for Offline Reinforcement Learning

    arXiv:2609.24489v2 Announce Type: replace-cross Abstract: Offline reinforcement learning (RL) typically trains a critic by minimizing a regression loss against bootstrapped value targets stabilized by target networks with exponential moving average (EMA) updates. Multi-step targets incorporate behavior-policy actions and therefore require off-policy correction. We instead impose in-sample Bellman optimality on the critic through inequality constraints. We formulate the Lifted Bellman Linear…

39. [39] 2026-09-28 Stepwise Intrinsic Rewards for Reasoning in Large Language Models

    arXiv:2602.01034v2 Announce Type: replace Abstract: Reinforcement learning (RL) has become a widely used paradigm for improving the reasoning abilities of large language models (LLMs) and Vision-language models (VLMs). Sparse binary outcome rewards, however, score only final correctness and cannot identify which intermediate steps contributed to it; in multimodal tasks, they may also reward answers driven by linguistic priors rather than…

40. [40] 2026-09-28 Different Corruptions, Different Signals: Uncertainty and Loss in Federated Data Quality

    arXiv:2609.31454v1 Announce Type: cross Abstract: Federated learning (FL) data corruption can affect either inputs or labels, but it remains unclear whether input-conditional uncertainty and prediction-label loss expose these corruption modes equally. This paper compares two corruption-detection signals in FL: input-conditional uncertainty and prediction-label loss. The uncertainty signal is characterised using a learned aleatoric variance estimate together with Monte Carlo (MC)…

41. [41] 2026-09-28 PUBG Ally: A Conversational Embodied Agent as an AI Teammate

    arXiv:2609.29837v2 Announce Type: replace Abstract: We introduce PUBG Ally, an embodied agent for PUBG: BATTLEGROUNDS that can reason, act autonomously, and play alongside players as a voice-enabled teammate. Building such a teammate requires combining two difficult capabilities: it must perceive and respond to a constantly changing game world under strict latency constraints while interacting naturally with players, keeping its speech…

42. [42] 2026-09-28 Selective Off-Policy Reference Tuning with Plan Guidance

    arXiv:2605.11505v3 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards helps reasoning, but GRPO-style methods stall on hard prompts where all sampled rollouts fail. SORT adds a repair update for those failures without changing rollout generation: it derives a plan from the reference solution, compares token probabilities with and without that plan, and gives higher weight to tokens that become…

Sources

1. [1] One year in: How Microsoft Research Asia – Singapore is advancing research, partnership and talent for real-world impact
2. [2] New formulation helps RNA vaccines withstand high temperatures
3. [3] All In Good Time: Causality-Aware Framework for LLM-Based Simultaneous Speech-to-Speech Translation
4. [4] Same Text, Different Numbers: The Divergence of LLM-Based Measures
5. [5] A Unified Account of Concepts and Chunks
6. [6] Thinking Less to Simulate Better: Intuitive Prompting Improves LLM Agents Simulating Individual Social Media Reactions, Including Unfamiliar Content
7. [7] The Communication Map of a Transformer
8. [8] A Benchmark Framework for Screening Automation in Systematic Reviews
9. [9] Not All Memories Are Equal: Hierarchical Collaborative Memory for Validity-Aware Retrieval in LLM Agents
10. [10] Auditing and Repairing LLM-as-Judge Failures in a Production Text-to-SQL Pipeline
11. [11] A Survey on Fake Review Detection: From Pre-trained Language Models to Large Language Models
12. [12] SlideLab: Audience-Centered Scientific Slide Generation and Evaluation
13. [13] Cartograph: Federated Tool Discovery with Operator-Attested Retrieval for AI Agents
14. [14] Bootstrapping Conversational Recommendation Agents At Spotify: Synthetic Data Generation and Self-Improvement Loops
15. [15] SignTrace: Describe a Sign, Find the Word
16. [16] The Right Information Extraction Pipeline Depends on the Document: Accuracy-Energy Trade-offs for Small, Local Models
17. [17] Symbiotic Architecture for Post-Hoc Audio Extension of Frozen Language Models
18. [18] VLAA-GUI: Knowing When to Stop, Recover, and Search, A Modular Framework for GUI Automation
19. [19] Learning to Stop without Learning to Stop: Self-Supervised Confidence Training Improves Reasoning Efficiency
20. [20] ArGuard Shared Task: Harmful Content Detection in Arabic Memes and LLM Prompts
21. [21] Attention-Discounted Adaptive Sampler for Masked Diffusion Language Models
22. [22] FlyAOC: Evaluating Agentic Ontology Curation of Drosophila Scientific Knowledge Bases
23. [23] Rethinking Data Quality for AI-Driven Systems: Evidence from Practitioner Interviews
24. [24] BioEVAL: A global, multi-institutional benchmark of large language and multimodal models for bioengineering
25. [25] PolyChirp: Multi-Species Birdsong Classification Using TinyML on Low-Power Acoustic Sensors
26. [26] Do LLMs Understand Context? A Knowledge Graph-Based Evaluation Framework
27. [27] An AI Agent Execution Environment to Safeguard User Data
28. [28] Pretrained ASR Pseudo-labeling for Noisy Police Audio
29. [29] Consist-Retinex: One-Step Noise-Emphasized Consistency Training Accelerates High-Quality Retinex Enhancement
30. [30] When Is a Multi-Agent Code Judge Actually Grounded? Two Label-Free Measurements, and a Judge That Declines to Guess
31. [31] Bridging LLM Agents and Data Spaces: An Architectural Mediation Approach using the Model Context Protocol
32. [32] Stealth Apart, Harm Together: Skill Cascading Attacks on Skill-Based Agent Systems
33. [33] Predicting Transmembrane Protein Topology from 3D Structure
34. [34] A Synthetic Ground-Truth Framework for the Evaluation of Explainable AI Methods
35. [35] Policy Regret for Embedding Model Routing: Contextual Bandits with Low-Rank Experts
36. [36] Spectral Feedback for Test-Time Alignment of Protein Diffusion Models
37. [37] UniAR: A Unified Framework for Autism Recognition Enhanced by Multi-View Prompt Learning
38. [38] Lifted Bellman Linear Programming for Offline Reinforcement Learning
39. [39] Stepwise Intrinsic Rewards for Reasoning in Large Language Models
40. [40] Different Corruptions, Different Signals: Uncertainty and Loss in Federated Data Quality
41. [41] PUBG Ally: A Conversational Embodied Agent as an AI Teammate
42. [42] Selective Off-Policy Reference Tuning with Plan Guidance