❌

Normal view

Received — 10 September 2026 ⏭

Distilling Image Prototypes for Guided Test-Time Adaptation

✇cs.AI, q-bio.NC updates on arXiv.org
By: Liwen Wang · Xingbo Dong · Iman Yi Liao · Deyin Liu · Massimo Tistarelli · Lin Yuanbo Wu · Zhe Jin
10 September 2026 at 12:00
arXiv:2609.09737v1 Announce Type: cross Abstract: Test-Time Adaptation (TTA) enhances the robustness of models against distribution shifts but faces two critical challenges: error accumulation from noisy pseudo-labels and catastrophic forgetting of source knowledge. Uncertainty-based approaches designed to mitigate error accumulation often yield overconfident or computationally expensive estimates, while strategies intended to prevent forgetting via prototype replay rely on static representations that easily become misaligned as the model adapts. To address these issues, this paper proposes a novel framework, Distilling Image Prototype for Guided Test-Time Adaptation (DIPTTA). The core of the proposed approach is the introduction of a Distill Image Prototype (DIP), a compact set of synthetic images that serves as a dynamic and regenerative anchor of source knowledge. This prototype enables a dynamic feature replay mechanism that continuously generates feature prototypes aligned with the current state of the model, thus effectively preventing catastrophic forgetting. Furthermore, the DIP anchors a source-calibrated uncertainty estimation method, which provides a less biased measure of sample reliability by leveraging stable source knowledge, thereby robustly suppressing error accumulation. Extensive experiments on multiple benchmarks demonstrate that DIPTTA significantly outperforms state-of-the-art methods, particularly under severe domain shifts. The source code is available at https://github.com/LiwenWang919/DIPTTA.

FlowCPO: A Unified Divergence View of Preference Alignment for Flow Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yansen Han · Shengyi Liao · Peng Sun · Deyuan Liu · Yuanxing Zhang · Pengfei Wan · Tao Lin
10 September 2026 at 12:00
arXiv:2609.09905v1 Announce Type: cross Abstract: Preference alignment for flow and diffusion models now spans online reinforcement learning and offline preference optimization, but the relation between these methods remains unclear. In particular, existing forward-process alignment methods require fresh samples from the current model, while offline methods based on fixed preference pairs rely primarily on positive-only fine-tuning or DPO-style likelihood-ratio surrogates. We organize these approaches through a divergence-based framework and introduce FlowCPO, an offline forward-KL objective that uses both preferred and dispreferred samples without online rollouts. For linear interpolation, we show under explicit regularity conditions that the forward-KL objective is bounded by a contrastive flow matching loss, yielding a tractable surrogate on fixed data. We further show that this loss is nonnegative, whereas the signed regression loss of simplified FlowDPO can be unbounded below. In the in-domain setting, FlowCPO achieves higher mean GenEval and OCR scores than the evaluated baselines, reaching 0.84 and 0.87 versus 0.81 and 0.74 for FlowDPO at CFG 3.0. In the out-of-domain setting, the results are mixed, with the best GenEval result but lower reward scores than RFT on several metrics.

Semigroup-JEPA: Latent Dynamics Consistency for Zero-Shot Physics Generalization

✇cs.AI, q-bio.NC updates on arXiv.org
By: Andy Zeyi Liu · Haoran Sun · Lucas Baker · Randall Balestriero · John Sous
10 September 2026 at 12:00
arXiv:2609.10464v1 Announce Type: cross Abstract: Joint-Embedding Predictive Architecture (JEPA) world models learn a compact latent representation of the world that supports prediction and planning, but their capability to learn physics and generate physically realistic dynamics remains hitherto untested. In this work, we introduce SemiGroup-JEPA (SG-JEPA), which extends the LeWorldModel framework by supplying the parameter governing the physics to the temporal model via action-conditioning and jointly training an encoder and predictor through an autoregressive latent rollout. To evaluate the model's ability to generalize out of distribution, we design dynamical tasks under different gravitational fields that, despite obeying the same physical law, exhibit qualitatively different dynamics, ranging from floating motion in weak gravitational fields to rapid bouncing in strong ones. In contrast to DINO-WM, SG-JEPA reduces open-loop prediction error by up to 2 times on two-dimensional datasets, and increases control success rate up to 2.5 times for three-dimensional robotic datasets, for which we train independent diffusion policies. To explain this advantage, we develop a linear feature model that separates local law-conditioned error from its recursive amplification under rollout. Guided by this model, we find that back-propagating the multi-step rollout loss into the representation trains the encoder to keep the features that the predictor can carry forward, and that those are the features the dynamics depend on, so most of the gain comes from the encoder learning better features rather than from the predictor learning better dynamics. See project page at https://sg-jepa.github.io.

MADS: Multi-Agent Dialogue Simulation for Diverse Persuasion Data Generation

✇cs.AI, q-bio.NC updates on arXiv.org
By: Mingjin Li · Yu Liu · Huayi Liu · Xiang Ye · Chao Jiang · Hongguang Zhang · Yu Ruan
10 September 2026 at 12:00
arXiv:2510.05124v3 Announce Type: replace-cross Abstract: We propose MADS (Multi-Agent Dialogue Simulation), a scalable framework for generating persuasive multi-turn dialogues via agent self-play. MADS employs three coordinated agents: User Agents designed to simulate diverse persona-driven behaviors by leveraging personality signifiers such as Zodiac Signs and MBTI types, a Dialog Agent executing task-oriented persuasion strategies and an Optimization Agent evaluating and refining dialogue outcomes. We further validate its effectiveness through users' Chain-of-Attitude (CoA) modeling and dedicated LLMs' persuasion assessment. This approach enables low-cost generation of training data without human annotation, addressing key industry challenges such as lack of user data, cold-start evaluation difficulties, and prompt inefficiency. Applied to a real-world marketing scenario, MADS significantly improved the persuasion capacity of small LLMs, increasing the organic traffic conversion rate by 22.4% (from 1.83% to 2.24%) , demonstrating clear business value.

Transmembrane glycoprotein BSG serves a dual role as a prognostic and immunological modulator in the tumor microenvironment of lung adenocarcinoma

✇Omics In Lung
By: Xiaofei Zhang · Qiliang Peng · Wenlang Zhu · Jianguo Zhang · Zhengfei Zhu · Kuaile Zhao · Peiyi Liu · Weixin Zhao
8 September 2026 at 18:00

Transl Oncol. 2026 Sep 8;73:102990. doi: 10.1016/j.tranon.2026.102990. Online ahead of print.

ABSTRACT

BACKGROUND: Lung adenocarcinoma (LUAD) is a predominant and lethal subtype of non-small cell lung cancer, with a lack of reliable prognostic biomarkers to guide clinical management. Basigin (BSG) has been implicated in tumor progression across multiple cancers, yet its expression pattern, prognostic significance, and underlying mechanisms in LUAD remain incompletely elucidated.

METHODS: We integrated multi-omics data from TCGA, GTEx, CCLE, and GEO databases to analyze BSG expression profiles. Clinical correlations were assessed via Kruskal-Wallis tests. Prognostic value was determined using Kaplan-Meier survival analysis, univariate/multivariate Cox regression, and nomogram construction with calibration curves. Functional enrichment (GO/KEGG) and immune infiltration analyses were performed to explore BSG-related mechanisms, followed by immunohistochemical (IHC) validation in A549 cells and clinical LUAD tissue microarrays.

RESULTS: BSG was significantly upregulated in LUAD tissues versus normal/paired adjacent tissues, correlating with advanced T/N/pathologic stages. High BSG expression predicted worse survival outcomes in TCGA-LUAD, which was validated in GEO datasets. Multivariate Cox regression identified BSG as an independent prognostic factor, with a well-calibrated nomogram for survival prediction. Functional exploration indicated that BSG mainly participated in tumor-associated and immunological pathways. Immune infiltration analysis indicated that BSG was significantly correlated with the infiltration of various immune cells. Moreover, BSG exhibited a strong association with immune checkpoint proteins, chemokines, chemokine receptors, and MHC genes. IHC further confirmed its cytoplasmic/membranous localization and prognostic relevance.

CONCLUSION: BSG serves as an independent prognostic biomarker and potential therapeutic target in LUAD, shedding light on its regulatory roles in tumor progression and immune microenvironment remodeling.

PMID:42710246 | DOI:10.1016/j.tranon.2026.102990

Transmembrane glycoprotein BSG serves a dual role as a prognostic and immunological modulator in the tumor microenvironment of lung adenocarcinoma

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Xiaofei Zhang · Qiliang Peng · Wenlang Zhu · Jianguo Zhang · Zhengfei Zhu · Kuaile Zhao · Peiyi Liu · Weixin Zhao
8 September 2026 at 18:00

Transl Oncol. 2026 Sep 8;73:102990. doi: 10.1016/j.tranon.2026.102990. Online ahead of print.

ABSTRACT

BACKGROUND: Lung adenocarcinoma (LUAD) is a predominant and lethal subtype of non-small cell lung cancer, with a lack of reliable prognostic biomarkers to guide clinical management. Basigin (BSG) has been implicated in tumor progression across multiple cancers, yet its expression pattern, prognostic significance, and underlying mechanisms in LUAD remain incompletely elucidated.

METHODS: We integrated multi-omics data from TCGA, GTEx, CCLE, and GEO databases to analyze BSG expression profiles. Clinical correlations were assessed via Kruskal-Wallis tests. Prognostic value was determined using Kaplan-Meier survival analysis, univariate/multivariate Cox regression, and nomogram construction with calibration curves. Functional enrichment (GO/KEGG) and immune infiltration analyses were performed to explore BSG-related mechanisms, followed by immunohistochemical (IHC) validation in A549 cells and clinical LUAD tissue microarrays.

RESULTS: BSG was significantly upregulated in LUAD tissues versus normal/paired adjacent tissues, correlating with advanced T/N/pathologic stages. High BSG expression predicted worse survival outcomes in TCGA-LUAD, which was validated in GEO datasets. Multivariate Cox regression identified BSG as an independent prognostic factor, with a well-calibrated nomogram for survival prediction. Functional exploration indicated that BSG mainly participated in tumor-associated and immunological pathways. Immune infiltration analysis indicated that BSG was significantly correlated with the infiltration of various immune cells. Moreover, BSG exhibited a strong association with immune checkpoint proteins, chemokines, chemokine receptors, and MHC genes. IHC further confirmed its cytoplasmic/membranous localization and prognostic relevance.

CONCLUSION: BSG serves as an independent prognostic biomarker and potential therapeutic target in LUAD, shedding light on its regulatory roles in tumor progression and immune microenvironment remodeling.

PMID:42710246 | DOI:10.1016/j.tranon.2026.102990

ITGA5 promotes homologous recombination mediated radioresistance in esophageal squamous cell carcinoma by upregulating RAD51AP1 expression

✇Oncogene - Issue - nature.com science feeds
By: Xinyi Huang · Jiayi Liu · Jiao Ren · Jupeng Yuan · Wen Huo · Zengfu Zhang · Minxin Chen · Baoqing Tian · Dawei Chen
4 September 2026 at 08:00

Oncogene, Published online: 04 September 2026; doi:10.1038/s41388-026-03966-8

ITGA5 promotes homologous recombination mediated radioresistance in esophageal squamous cell carcinoma by upregulating RAD51AP1 expression

Genomics and social practices at Mogou and other Gansu sites during prehistoric trans-Eurasian exchange

✇Cell
By: Han Shi · E. Andrew Bennett · Guoke Chen · Hui Wang · Tianyi Wang · Peng Cao · Ruilin Mao · Fan Bai · Qingyan Dai · Feng Liu · Xiaotian Feng · Yongan Wang · Chong Yao · Jing Wu · Wenjun Wang · Ming Zhang · Tianxiang Liu · Jingkun Ran · Yue-Chen Liu · Yuan Liu · Dan Ju · Chuyan Li · Hongwei Hou · Xinyi Liu · Qiaomei Fu
3 September 2026 at 08:00
Ancient DNA from 149 individuals at 11 sites in Gansu, China, dated to around 4,700–3,000 years ago, reveals human population history during early transcontinental exchanges of agriculture and technology, as well as contemporary social practices, at the large Mogou cemetery.
Received — 27 May 2026 ⏭

GlobalDentBench: A Multinational Benchmark for Evaluating LLM Clinical Reasoning in Dentistry with Expert Calibration

✇cs.AI, q-bio.NC updates on arXiv.org
By: Junjie Zhao · Jingyi Liang · Zhenyang Cai · Jiaming Zhang · Zhenwei Wen · Shuzhi Deng · Wenjing Yi · Chunfeng Luo · Hexian Zhang · Junying Chen · Tianrui Liu · Zhuhui Bai · Zixu Zhang · Pradeep Singh · Xiang Liu · Jianquan Li · Nhan L Tran · Falk Schwendicke · Zuolin Jin · Lijian Jin · Liangyi Chen · Wei-fa Yang · Benyou Wang · Junwen Wang · Shan Jiang
26 May 2026 at 12:00
arXiv:2605.24636v2 Announce Type: new Abstract: While large language models (LLMs) hold transformative potential for medicine, their reasoning robustness and safety in real-world clinical scenarios remain critically underexplored, particularly in dentistry. Here we introduce GlobalDentBench, the first multinational dental benchmark, featuring a taxonomy that encompasses 14 dental specialties across 88 countries and regions spanning six continents. The benchmark comprises 8,978 expert-validated questions across three formats (multiple-choice, short-answer, and case-based questions) and assesses three progressive reasoning levels: knowledge recall (L1), routine reasoning (L2), and individualized reasoning (L3). To ensure data quality, the automated construction framework was calibrated by six senior dentists, achieving expert agreement rates of 99.98% for multiple-choice and short-answer questions and 96.78% for the more complex case-based questions. Evaluation of 12 frontier LLMs on GlobalDentBench revealed a sharp, stepwise performance degradation with increasing reasoning complexity. Specifically, accuracy plummeted from 81.34% on multiple-choice to 64.53% on short-answer and 22.34% on case-based questions, while declining markedly from 74.01% at L1 to 55.64% at L2 and 35.71% at L3. More critically, risk analysis of real-world dental cases demonstrated an alarming overall unsafe rate of 31.01% in LLM-generated clinical recommendations, with 4.51% posing risks of irreversible patient harm and risks particularly pronounced in specialties such as orthodontics. These findings expose fundamental limitations in the medical reasoning and safety of current LLMs. Consequently, GlobalDentBench provides a scalable foundation for trustworthy clinical AI evaluation, underscoring the urgent need for rigorous validation before the safe deployment of these models in healthcare.

DarkForest: Less Talk, Higher Accuracy for Multi-Agent LLMs

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yi Li · Songtao Wei · Dongming Jiang · Zhichun Guo · Qiannan Li · Bingzhe Li
26 May 2026 at 12:00
arXiv:2605.25188v1 Announce Type: new Abstract: Multi-agent LLM systems improve reasoning by combining outputs from multiple agents, but interaction-heavy methods can introduce error propagation and high communication overhead. When agents exchange raw responses or reasoning traces, incorrect intermediate reasoning may be adopted and amplified, leading to confident but wrong consensus; multi-round communication also increases token consumption, latency, and inference cost. In this paper, we propose a controlled-communication coordination framework named DarkForest. DarkForest first keeps agents independent, so each agent produces an answer without seeing the others' outputs. It then parses the raw responses into structured candidate records, groups semantically equivalent candidates into clusters, and estimates a calibrated belief distribution over these clusters using agent reliability, confidence, parse quality, support-pattern reliability, and independence corrections. A coordinator receives only policy-permitted evidence from this belief state with controlled communication. Experiments on six reasoning benchmarks show that DarkForest achieves leading overall quality, improves the strongest baseline by up to 30.7\% on benchmark metrics, and reduces token consumption by up to $6.5\times$ compared with communication-heavy baselines.

FrontierOR: Benchmarking LLMs' Capacity for Efficient Algorithm Design in Large-Scale Optimization

✇cs.AI, q-bio.NC updates on arXiv.org
By: Minwei Kong · Chonghe Jiang · Ao Qu · Wenbin Ouyang · Zhaoming Zeng · Xiaotong Guo · Zhekai Li · Junyi Li · Yi Fan · Xinshou Zheng · Xi Jing · Yikai Zhang · Zhiwei Liang · Seonghoo Kim · Runqing Yang · Zijian Zhou · Sirui Li · Han Zheng · Wangyang Ying · Ou Zheng · Chonghuan Wang · Jinglong Zhao · Hanzhang Qin · Cathy Wu · Paul Pu Liang · Jinhua Zhao · Hai Wang
26 May 2026 at 12:00
arXiv:2605.25246v2 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for optimization modeling and solver-code generation, yet practical operations research and optimization problems often require a harder capability: designing scalable algorithms that exploit problem structure and outperform direct formulation-and-solve baselines. Existing benchmarks are limited to small or simplified examples far below real-world scale and complexity. We introduce FrontierOR, among the first benchmarks to systematically evaluate LLM-based efficient algorithm design for realistic large-scale optimization problems. FrontierOR includes 180 tasks derived from methodologically diverse papers published in top-tier operations research venues, each with standardized instances and a hidden, expert-verified evaluation suite. We evaluate seven LLMs spanning frontier, cost-effective, and open-source models both in one-shot and test-time evolution settings. The results reveal that frontier models still struggle to move from executable formulations to efficient optimization algorithms: the strongest one-shot model outperforms Gurobi in only 31% of cases in both solution quality and computational efficiency, and even strong coding agents with test-time evolution achieve only 50% on selected hard tasks. FrontierOR establishes a practical evaluation platform for LLM-based optimization algorithm design, which enables future LLMs and agents to be systematically tested on whether they can move beyond correct formulation toward a feasible, high-quality, and efficient algorithm.

Towards end-to-end LLM-based censoring-aware survival analysis

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yishu Wei · Hexin Dong · Yi Lin · Jiahe Qian · Yi Liu · Yifan Peng
26 May 2026 at 12:00
arXiv:2605.25399v1 Announce Type: new Abstract: Objective: Survival analysis is central to medical prediction, yet large language models (LLMs) are rarely used as end-to-end survival models because censoring prevents straightforward supervised fine-tuning. Here we present LLMSurvival, a framework that enables censoring-aware survival analysis with unmodified LLMs operating directly on tabular clinical data. Materials and Methods: LLMSurvival reformulates time-to-event prediction as pairwise ranking among comparable subjects, and derives test-time risk by aggregating comparisons against anchor individuals from the training cohort. Results: Across two clinical tasks (ICU mortality prediction in MIMIC-IV and fragility fracture prediction in a NewYork-Presbyterian/Weill Cornell Medicine cohort), LLMSurvival improves overall concordance over Cox proportional hazards modeling by 3.1% for ICU mortality and 0.5% for fracture risk, 2.1% on average for ICU mortality and 2.8% for fracture risk over three established deep learning survival models. Discussion: The results show that survival modeling with censoring can be made compatible with LLM fine-tuning through comparison-based reformulation. The framework demonstrates high portability and superior performance over expert curated scores like SAPS-II and FRAX scores across diverse clinical context. Furthermore, the framework supports local deployment, as compact, publicly available base models provide sufficient performance. Conclusion: The LLMSurvival framework serves as a proof of concept for an integrated, censoring-conscious approach to survival analysis via LLMs.

AgentHijack: Benchmarking Computer Use Agent Robustness to Common Environment Corruptions

✇cs.AI, q-bio.NC updates on arXiv.org
By: Jingwei Sun · Jianing Zhu · Yuanyi Li · Tongliang Liu · Xia HU · Bo Han
26 May 2026 at 12:00
arXiv:2605.25707v1 Announce Type: new Abstract: Autonomous computer use agents that powered by multimodal large language models (MLLMs) are emerging as capable assistants for completing complex digital workflows. However, real-world execution environments are far from ideal: pop-ups, resolution changes, and competing applications frequently interfere with agent perception and control. We introduce AgentHijack, a benchmark designed to evaluate the robustness of computer-use agents under common corruptions, where the uncertainties in dynamic environment disrupt the execution flow without direct adversarial intent. Specifically, AgentHijack introduces 9 configurable common corruptions to replicate realistic imperfect scenarios. We evaluate a variety of desktop tasks that utilize MLLM-based agents and discover that even minor instances of corruption can result in substantial performance degradation, which emphasizes the fragility of agents and underscores the necessity of robustness evaluation. Afterward, we propose AgentHijack-Agent, a framework that integrates an action generator with enhanced grounding capabilities and an onlooker responsible for behavior summarization and environment checking. Extensive experiments validate its effectiveness. Our code, environment, baseline models and data are publicly available at: https://AgentHijack.github.io.

Agent Learning via Early Experience

✇cs.AI, q-bio.NC updates on arXiv.org
By: Kai Zhang · Xiangchao Chen · Bo Liu · Tianci Xue · Zeyi Liao · Zhihan Liu · Xiyao Wang · Yuting Ning · Zhaorun Chen · Xiaohan Fu · Jian Xie · Yuxuan Sun · Boyu Gou · Qi Qi · Zihang Meng · Jianwei Yang · Ning Zhang · Xian Li · Ashish Shah · Dat Huynh · Hengduo Li · Zi Yang · Sara Cao · Lawrence Jang · Shuyan Zhou · Jiacheng Zhu · Huan Sun · Jason Weston · Yu Su · Yifan Wu
26 May 2026 at 12:00
arXiv:2510.08558v3 Announce Type: replace Abstract: A long-term goal of language agents is to learn and improve through their own experience, ultimately outperforming humans in complex, real-world tasks. However, training agents from experience data with reinforcement learning remains difficult in many environments, which either lack verifiable rewards (e.g., websites) or require inefficient long-horizon rollouts (e.g., multi-turn tool use). As a result, most current agents rely on supervised fine-tuning on expert data, which is challenging to scale and generalizes poorly. This limitation stems from the nature of expert demonstrations: they capture only a narrow range of scenarios, and expose the agent to limited environment diversity. We address this limitation with a middle-ground paradigm we call early experience: interaction data generated by the agent's own actions, where the resulting future states serve as supervision without reward signals. Within this paradigm, we study two strategies of using such data: (1) implicit world modeling, which uses collected states to ground the policy in environment dynamics; and (2) self-reflection, where the agent learns from its suboptimal actions to improve reasoning and decision-making. Evaluation across eight diverse environments and multiple model families shows that our approaches consistently improve effectiveness and out-of-domain generalization, highlighting the value of early experience. Moreover, in environments with verifiable rewards, our results provide promising signals that early experience offers a strong foundation for subsequent reinforcement learning, making it a practical bridge between imitation learning and fully experience-driven agents.

Agent World Model: Infinity Synthetic Environments for Agentic Reinforcement Learning

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zhaoyang Wang · Canwen Xu · Boyi Liu · Yite Wang · Siwei Han · Zhewei Yao · Huaxiu Yao · Yuxiong He
26 May 2026 at 12:00
arXiv:2602.10090v3 Announce Type: replace Abstract: Recent advances in large language model (LLM) have empowered autonomous agents to perform multi-turn interactions with tools and environments. However, scaling such agent training is limited by the lack of diverse and reliable environments. In this paper, we propose Agent World Model (AWM), a fully synthetic environment generation pipeline. Using this pipeline, we scale to 1,000 environments covering everyday scenarios, in which agents can interact with rich toolsets and obtain high-quality observations. Notably, these environments are code-driven and backed by databases, providing more reliable and consistent state transitions than environments simulated by LLMs. Moreover, they enable more efficient agent interaction compared with collecting trajectories from realistic environments. To demonstrate the effectiveness of this resource, we perform large-scale reinforcement learning for multi-turn tool-use agents. Thanks to the fully executable environments and accessible database states, we can also design reliable reward functions. Experiments on three benchmarks show that training exclusively in synthetic environments, rather than benchmark-specific ones, yields strong out-of-distribution generalization. The code is available at https://github.com/Snowflake-Labs/agent-world-model.

Reliable AI Needs to Externalize Implicit Knowledge: A Human-AI Collaboration Perspective

✇cs.AI, q-bio.NC updates on arXiv.org
By: Hengyu Liu · Tianyi Li · Zhihong Cui · Yushuai Li · Zhangkai Wu · Torben Bach Pedersen · Kristian Torp · Christian S. Jensen
26 May 2026 at 12:00
arXiv:2605.02010v2 Announce Type: replace Abstract: This position paper argues that reliable AI requires infrastructure for human validation of implicit knowledge. AI learns from both explicit knowledge (papers, documentation, structured databases) and implicit knowledge (reasoning patterns, debugging processes, intermediate steps). Implicit knowledge remains unexternalized because documentation cost exceeds perceived value -- yet AI learns from it indiscriminately, acquiring both beneficial patterns and harmful biases. Current reliability methods can only verify explicit knowledge against sources, creating a fundamental gap: the most valuable AI capabilities (reasoning, judgment, intuition) are precisely those we cannot verify. We propose Knowledge Objects (KOs) -- structured artifacts that externalize implicit knowledge into forms humans can inspect, verify, and endorse. KOs transform verification economics: what was previously too costly to verify becomes feasible, enabling accumulated human validation to improve reliability over time.

Bridging Evolutionary Algorithms and Reinforcement Learning: A Comprehensive Survey on Hybrid Algorithms

✇cs.AI, q-bio.NC updates on arXiv.org
By: Pengyi Li · Jianye Hao · Hongyao Tang · Xian Fu · Yan Zheng · Ke Tang
26 May 2026 at 12:00
arXiv:2401.11963v5 Announce Type: replace-cross Abstract: Evolutionary Reinforcement Learning (ERL), which integrates Evolutionary Algorithms (EAs) and Reinforcement Learning (RL) for optimization, has demonstrated remarkable performance advancements. By fusing both approaches, ERL has emerged as a promising research direction. This survey offers a comprehensive overview of the diverse research branches in ERL. Specifically, we systematically summarize recent advancements in related algorithms and identify three primary research directions: EA-assisted Optimization of RL, RL-assisted Optimization of EA, and synergistic optimization of EA and RL. Following that, we conduct an in-depth analysis of each research direction, organizing multiple research branches. We elucidate the problems that each branch aims to tackle and how the integration of EAs and RL addresses these challenges. In conclusion, we discuss potential challenges and prospective future research directions across various research directions. To facilitate researchers in delving into ERL, we organize the algorithms and codes involved on https://github.com/yeshenpy/Awesome-Evolutionary-Reinforcement-Learning.

Reward-free Alignment for Conflicting Objectives

✇cs.AI, q-bio.NC updates on arXiv.org
By: Peter Chen · Xiaopeng Li · Xi Chen · Tianyi Lin
26 May 2026 at 12:00
arXiv:2602.02495v3 Announce Type: replace-cross Abstract: Direct alignment methods are increasingly used to align large language models (LLMs) with human preferences. However, many real-world alignment problems involve multiple conflicting objectives, where naive aggregation of preferences can lead to unstable training and poor trade-offs. In particular, weighted loss methods may fail to identify update directions that simultaneously improve all objectives, and existing multi-objective approaches often rely on explicit reward models, introducing additional complexity and distorting user-specified preferences. The contributions of this paper are two-fold. First, we propose a Reward-free Alignment framework for Conflicted Objectives (RACO) that directly leverages pairwise preference data and resolves gradient conflicts via a novel clipped variant of conflict-averse gradient descent. We provide convergence guarantees to Pareto-critical points that respect user-specified objective weights, and further show that clipping can strictly improve convergence rate in the two-objective setting. Second, we improve our method using some heuristics and conduct experiments to demonstrate the compatibility of the proposed framework for LLM alignment. Both qualitative and quantitative evaluations on multi-objective summarization and safety alignment tasks across multiple LLM families (Qwen 3, Llama 3, Gemma 3) show that our method consistently achieves better Pareto trade-offs compared to existing multi-objective alignment baselines.

SPA-Cache: Singular Proxies for Adaptive Caching in Diffusion Language Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Wenhao Sun · Rong-Cheng Tu · Yifu Ding · Zhao Jin · Jingyi Liao · Yongcheng Jing · Dacheng Tao
26 May 2026 at 12:00
arXiv:2602.02544v2 Announce Type: replace-cross Abstract: While Diffusion Language Models (DLMs) offer a flexible, arbitrary-order alternative to the autoregressive paradigm, their non-causal nature precludes standard KV caching, forcing costly hidden state recomputation at every decoding step. Existing DLM caching approaches reduce this cost by selective hidden state updates; however, they are still limited by (i) costly token-wise update identification heuristics and (ii) rigid, uniform budget allocation that fails to account for heterogeneous hidden state dynamics. To address these challenges, we present SPA-Cache that jointly optimizes update identification and budget allocation in DLM cache. First, we derive a low-dimensional singular proxy that enables the identification of update-critical tokens in a low-dimensional subspace, substantially reducing the overhead of update identification. Second, we introduce an adaptive strategy that allocates fewer updates to stable layers without degrading generation quality. Together, these contributions significantly improve the efficiency of DLMs, yielding up to an $8\times$ throughput improvement over vanilla decoding and a $2$--$4\times$ speedup over existing caching baselines.
❌