❌

Normal view

Received — 10 September 2026 ⏭

Hierarchical and Permutation-Invariant Feature Transformation Learning via Policy-Guided Embedding Search

✇cs.AI, q-bio.NC updates on arXiv.org
By: Rui Liu · Tao Zhe · Yanyong Huang · Sankha Narayan Guria · Xiao Luo · Wei Fan · Yanjie Fu · Dongjie Wang
10 September 2026 at 12:00
arXiv:2609.10225v1 Announce Type: cross Abstract: Feature transformation improves predictive performance on tabular data by constructing informative abstractions from raw features. Recent generative approaches encode transformation knowledge into continuous embedding spaces for efficient exploration of candidate strategies, but face three key limitations: (1) overlooking hierarchical relationships between low-level features, operations, and high-level abstractions; (2) enforcing order-sensitive embeddings on inherently permutation-invariant transformation sequences, thereby introducing systematic bias; and (3) relying on gradient-based search, which is ill-suited to non-convex transformation spaces. We propose a framework with two complementary components. First, a permutation-invariant hierarchical module captures interactions across features, operations, and abstraction levels, with a self-attention pooling mechanism that maps semantically equivalent structures to consistent embeddings aligned with downstream performance. Second, a policy-guided multi-objective reinforcement learning strategy initializes the search from empirically strong seeds and jointly optimizes predictive accuracy and transformation efficiency. Extensive experiments on diverse tabular benchmarks demonstrate the effectiveness and robustness of our framework against strong baselines. Our code and data are publicly available at: https://github.com/RayLiu1103/PHER.

City Editing: Hierarchical Agentic Execution for Dependency-Aware Urban Geospatial Modification

✇cs.AI, q-bio.NC updates on arXiv.org
By: Rui Liu · Steven Jige Quan · Zhong-Ren Peng · Zijun Yao · Han Wang · Pengyang Wang · Kunpeng Liu · Yanjie Fu · Dongjie Wang
10 September 2026 at 12:00
arXiv:2602.19326v3 Announce Type: replace-cross Abstract: Urban renewal requires incremental modifications to existing geospatial plans, yet manually updating complex layouts under spatial constraints is labor-intensive and error-prone. To tackle this, we propose CEAE, a hierarchical agentic framework that formulates urban renewal as machine-executable GeoJSON editing from natural-language instructions. CEAE decomposes instructions into hierarchical geometric intents, executing edits from coarse to fine while preserving spatial consistency through a self-reflective execution-validation loop. Experimental results show that CEAE outperforms baselines in execution validity, robustness, and geometric accuracy.

Early stage nonsmall cell lung cancer: Toward a risk-adaptive paradigm in the era of biologic precision

✇Omics In Lung
By: Chaoqi Zhang · Xuanyu Gu · Pan Wang · Dongyu Li · Guochao Zhang · Ting Wang · Peng Wu · Wei Rao · Shan Muhammad · Binghua Tan · Yun Che · Mobai Zhu · Qilin Qiao · Fanbo Lai · Bohui Zhao · Kailun Fei · Yue Wu · Zhihui Zhang · Min Dai · Wanqing Chen · Yawei Zhang · Nan Bi · Jie Wang · Jie He
9 September 2026 at 18:00

CA Cancer J Clin. 2026 Sep-Oct;76(5):e70100. doi: 10.3322/caac.70100.

ABSTRACT

The clinical landscape of early stage nonsmall cell lung cancer is at transformative crossroads. Driven by the widespread adoption of low-dose computed tomography screening, the frequent detection of ground-glass opacities, and a rising incidence among never-smokers, the diagnostic center of gravity has shifted toward earlier, potentially curable disease. This shift has been accompanied by equally important therapeutic advances, including parenchyma-sparing surgical techniques, minimally invasive platforms enhanced by digital navigation, and the transformative integration of perioperative immunotherapy and targeted agents. Concurrently, noninvasive monitoring approaches, such as liquid biopsy, have emerged as powerful tools to guide precision management. Despite this progress, substantial barriers to achieving a universal cure persist. Clinicians continue to face uncertainty in the management of ground-glass opacities, the anatomy-based TNM staging system fails to capture the biologic heterogeneity of early tumors, and global disparities in access to innovation remain unresolved. To address these challenges, the authors propose a shift toward a risk-adaptive management paradigm that harnesses artificial intelligence-driven analytics and multi-omics profiling to tailor treatment intensity according to each patient's biologic risk. Such an approach would enable appropriate escalation for high-risk individuals while permitting safe de-escalation for those at low risk. This holistic, lifespan-oriented strategy must be embraced to deliver equitable and durable cures for patients with early stage nonsmall cell lung cancer.

PMID:42713910 | PMC:PMC13555834 | DOI:10.3322/caac.70100

Early stage nonsmall cell lung cancer: Toward a risk-adaptive paradigm in the era of biologic precision

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Chaoqi Zhang · Xuanyu Gu · Pan Wang · Dongyu Li · Guochao Zhang · Ting Wang · Peng Wu · Wei Rao · Shan Muhammad · Binghua Tan · Yun Che · Mobai Zhu · Qilin Qiao · Fanbo Lai · Bohui Zhao · Kailun Fei · Yue Wu · Zhihui Zhang · Min Dai · Wanqing Chen · Yawei Zhang · Nan Bi · Jie Wang · Jie He
9 September 2026 at 18:00

CA Cancer J Clin. 2026 Sep-Oct;76(5):e70100. doi: 10.3322/caac.70100.

ABSTRACT

The clinical landscape of early stage nonsmall cell lung cancer is at transformative crossroads. Driven by the widespread adoption of low-dose computed tomography screening, the frequent detection of ground-glass opacities, and a rising incidence among never-smokers, the diagnostic center of gravity has shifted toward earlier, potentially curable disease. This shift has been accompanied by equally important therapeutic advances, including parenchyma-sparing surgical techniques, minimally invasive platforms enhanced by digital navigation, and the transformative integration of perioperative immunotherapy and targeted agents. Concurrently, noninvasive monitoring approaches, such as liquid biopsy, have emerged as powerful tools to guide precision management. Despite this progress, substantial barriers to achieving a universal cure persist. Clinicians continue to face uncertainty in the management of ground-glass opacities, the anatomy-based TNM staging system fails to capture the biologic heterogeneity of early tumors, and global disparities in access to innovation remain unresolved. To address these challenges, the authors propose a shift toward a risk-adaptive management paradigm that harnesses artificial intelligence-driven analytics and multi-omics profiling to tailor treatment intensity according to each patient's biologic risk. Such an approach would enable appropriate escalation for high-risk individuals while permitting safe de-escalation for those at low risk. This holistic, lifespan-oriented strategy must be embraced to deliver equitable and durable cures for patients with early stage nonsmall cell lung cancer.

PMID:42713910 | DOI:10.3322/caac.70100

Multi-omics integration identifies APOE as a metabolic regulator of macrophage-fibroblast crosstalk in idiopathic pulmonary fibrosis

✇(Multiomics OR Omics) AND (Pancreatic)
By: Bing Bai · Le Chang · Zhi Cheng · Yingjie Wang · Chenfeng Hua · Yonghai Feng · Wenfei Zhao
2 September 2026 at 18:00

Front Immunol. 2026 Aug 18;17:1904638. doi: 10.3389/fimmu.2026.1904638. eCollection 2026.

ABSTRACT

BACKGROUND: Aberrant tissue repair and relentless fibroblast activation are hallmark features of idiopathic pulmonary fibrosis (IPF). Although IPF and Alzheimer's disease (AD) share underlying aging-related pathologies, including immune and metabolic dysregulation, the putative genetic mechanisms linking AD susceptibility to pathogenic macrophage remodeling in the fibrotic niche are not fully established.

METHODS: We performed a two-sample Mendelian randomization (MR) analysis to assess the genetic association and potential causal relationship between AD and IPF. Shared hub genes were identified via protein interaction networks. To characterize macrophage heterogeneity and intercellular crosstalk within the IPF microenvironment, we interrogated scRNA-seq data (GSE122960) utilizing Monocle 3 and CellChat algorithms. The functional essentiality of APOE was evaluated bridging computational virtual knockout (scTenifoldKnk) with laboratory in vitro assays. Specifically, downstream transcriptomic shifts and fibroblast activation capacities were validated using APOE-silenced THP-1 macrophages and a Transwell co-culture model with MRC-5 cells.

RESULTS: MR estimates indicated that genetic liability to AD is associated with a lower risk of developing IPF. Integrated profiling identified the lipid-metabolism gene APOE as a central hub, specifically enriched in lung macrophages. Pseudotime modeling captured a pathogenic bifurcation in IPF, where macrophages evolve toward a terminal state marked by profound oxidative phosphorylation defects and massive SPP1 secretion. These SPP1+ macrophages primarily activate fibroblasts via CD44 and integrin signaling axes. Furthermore, both virtual simulations and in vitro THP-1 experiments demonstrated that loss of APOE function triggers the hyperactivation of complement (C1QA) and antigen-presentation (HLA-DR) pathways. Co-culture assays ultimately confirmed that APOE ablation in macrophages strongly exacerbates myofibroblast differentiation (elevated α-SMA and collagen I) in adjacent MRC-5 cells.

CONCLUSION: APOE functions as a vital metabolic barrier against pro-fibrotic macrophage polarization in the lung. Disruption of this specific lipid metabolic network is strongly associated with SPP1-driven fibroblast activation and local immune imbalance, providing a theoretical framework that strictly warrants future in vivo investigation to determine its clinical relevance.

PMID:42682427 | PMC:PMC13529525 | DOI:10.3389/fimmu.2026.1904638

Nitrogen dioxide exposure promotes CD8(+)T cell infiltration and contributes to increased susceptibility to ulcerative colitis: An integrative multi-omics, artificial intelligence, and mouse model study

✇(Multiomics OR Omics) AND (Pancreatic)
By: Zhijie Wang · Xin Chang · Tian Fu · Dan Xu · Yuanhong Ma · Hao Zhang · Hubin Xia · Xiaofeng Zhang · Yingdong Hou
1 September 2026 at 18:00

J Hazard Mater. 2026 Sep 15;516:143449. doi: 10.1016/j.jhazmat.2026.143449. Epub 2026 Aug 30.

ABSTRACT

The global incidence of ulcerative colitis (UC) has significantly increased in rapidly industrializing nations, with numerous studies highlighting environmental exposures, particularly nitrogen dioxide (NO2), as potential contributors to disease susceptibility. However, the clinical implications and molecular mechanisms linking NO2 exposure to UC susceptibility remain poorly understood. This study investigated the associations between NO2 and UC by integrating multi-omics data. We identified a CD8+ T cell subpopulation with a distinct phenotype characterized by perforin production, which potentially exacerbated colonic inflammation related to NO2 exposure. To validate this hypothesis, we established mouse models exposed to NO2, confirming increased CD8+ T cell infiltration and elevated perforin secretion through immunofluorescent (IF) staining. Employing artificial intelligence techniques, we identified Cell Division Cycle 25B (CDC25B) as a gene of interest correlated with putative NO2-related UC signatures. Finally, through molecular docking (MD) and molecular dynamics simulations (MDS), we identified ozanimod as one of several computationally nominated compounds associated with the CDC25B‑related network; however, none of these computational predictions were experimentally validated in the present study. Collectively, these findings suggest a correlative link between perforin or CD8+ T cell-associated colonic inflammation and NO2-associated UC susceptibility, and nominate CDC25B as a candidate gene for further investigation.

PMID:42679583 | DOI:10.1016/j.jhazmat.2026.143449

Ibrutinib-triggered matriptase maintains extracellular CD19 and limits antigen escape in B-cell malignancy

✇Cell Death Discovery nature.com science feeds
By: Xin-Jie Lu · Hsing-Fan Lai · Yu-Sin Hung · Yu-Jie Wang · Sheng-Cheng Wu · Jehng-Kang Wang · Yi-Ying Wu · Yi-Lin Chiu
31 August 2026 at 08:00

Cell Death Discovery, Published online: 31 August 2026; doi:10.1038/s41420-026-03306-5

Ibrutinib-triggered matriptase maintains extracellular CD19 and limits antigen escape in B-cell malignancy

Skin-innervating glutamatergic neurons modulate aging

✇Cell
By: Zhikai Wang · Xinping Jin · Yue Wu · Wenxiang Ding · Fengjiao Zhang · Yang Sun · Fangzhou Lou · Jing Bai · Xiuli Yang · Hong Zhou · Jihuan Liang · Xiaojie Cai · Yan Li · Xichen Zheng · Jiajia Han · Fang Tong · Qing Yang · Zhixiao Fang · Liuqi Zhao · Qingqing Shen · Zhenyao Xu · Siyu Deng · Li Fan · Jiajia Tong · Linhui Tong · Junjie Wang · Min Gao · Zilong Fang · Qun Li · Rui Chen · Lina Huang · Zhen Li · Yuqin Yang · Jing Zou · Ying Wu · Xiwen Zhang · Honglin Wang
17 July 2026 at 08:00
Within the skin, glutamatergic neurons expressing neurofilament heavy chain (Nefh) play a role in aging. Loss of Nefh during aging drives skin fibroblast senescence and collagen loss, whereas glutamate supplementation improves skin aging phenotypes.
Received — 27 May 2026 ⏭

Disentangled Double Machine Learning for Accurate Causal Effect Estimation

✇cs.AI, q-bio.NC updates on arXiv.org
By: Guodu Xiang · Kui Yu · Yujie Wang · Richang Hong · Fuyuan Cao · Jiye Liang
26 May 2026 at 12:00
arXiv:2605.24808v1 Announce Type: cross Abstract: Confounding bias is a key challenge in causal effect estimation from observational data. Double Machine Learning (DML) addresses this issue by estimating treatment and outcome nuisance functions, constructing treatment and outcome residuals, and estimating causal effects from the residuals. However, DML often produces biased and unstable estimates in highdimensional or finite-sample scenarios. One reason is that DML estimates nuisance functions using all covariates without disentangling distinct latent factors, resulting in unreliable nuisance function estimation. Another is that imprecise nuisance estimation further introduces residual dependence between the treatment residual and the remaining outcome error, undermining the accuracy of causal effect estimates. To address these issues, in this paper, we propose Disentangled Double Machine Learning (DDML), a novel algorithm that integrates two key strategies. First, a causal role disentanglement strategy decomposes covariates into confounders, treatment-specific factors, and outcomespecific factors for enabling reliable nuisance function estimation. And second, a residual dependence orthogonalization strategy mitigates residual dependence caused by nuisance estimation errors for enhancing the precision of causal effect estimates. Experimental results on synthetic, semi-synthetic, and real-world datasets demonstrate that DDML significantly outperforms 13 state-of-the-art baseline algorithms in both MAE and RMSE.

PageLLM: A Multi-Grained Reward Framework for Whole-Page Optimization with Large Language Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Xinyuan Wang · Liang Wu · Dongjie Wang · Yanjie Fu
26 May 2026 at 12:00
arXiv:2506.09084v2 Announce Type: replace-cross Abstract: Whole-page optimization (WPO) decides how search and recommendation results are surfaced to users, and large language models (LLMs) open a new route to it by treating page generation as sequence generation. Adapting LLMs to web-scale WPO, however, remains bottlenecked by the need for costly human annotations and by the mismatched granularity between page-level coherence and item-level placement. In this work we show that these two challenges are coupled: implicit user feedback alone suffices for alignment, provided the reward signal is decoupled into two complementary granularities. We propose PageLLM, a reward-based fine-tuning framework that (i) turns implicit feedback into four contrastive preference-pair families covering relevance, ranking, diversity, and redundancy, (ii) learns a coarse page-level reward and a fine item-level reward that captures engagement-sensitive position swaps, and (iii) combines both rewards in PPO-based RLHF over a pre-trained LLM. Extensive experiments on seven Amazon categories against eleven baselines show that neither reward alone is sufficient -- dropping the page-level or item-level signal reduces NDCG@100 by 17.8% and 15.2% respectively, whereas the joint reward improves NDCG@100 by up to 46.8%. Deployed in a 10M-user online A/B test, PageLLM raises GMV by 0.44% and click-through rate by 0.14%, confirming that multi-grained rewards from implicit feedback scale to production WPO. Code and data are available at an anonymized repository.

ChunkLLM: A Lightweight Pluggable Framework for Accelerating LLMs Inference

✇cs.AI, q-bio.NC updates on arXiv.org
By: Haojie Ouyang · Jianwei Lv · Lei Ren · Chen Wei · Xiaojie Wang · Fangxiang Feng
26 May 2026 at 12:00
arXiv:2510.02361v2 Announce Type: replace-cross Abstract: Transformer-based large models excel in natural language processing and computer vision, but face severe computational inefficiencies due to the self-attention's quadratic complexity with input tokens. Recently, researchers have proposed a series of methods based on block selection and compression to alleviate this problem, but they either have issues with semantic incompleteness or poor training-inference efficiency. To comprehensively address these challenges, we propose ChunkLLM, a lightweight and pluggable training framework. Specifically, we introduce two components: QK Adapter (Q-Adapter and K-Adapter) and Chunk Adapter. The former is attached to each Transformer layer, serving dual purposes of feature compression and chunk attention acquisition. The latter operates at the bottommost layer of the model, functioning to detect chunk boundaries by leveraging contextual semantic information. During the training phase, the parameters of the backbone remain frozen, with only the QK Adapter and Chunk Adapter undergoing training. Notably, we design an attention distillation method for training the QK Adapter, which enhances the recall rate of key chunks. During the inference phase, chunk selection is triggered exclusively when the current token is detected as a chunk boundary, thereby accelerating model inference. Experimental evaluations are conducted on a diverse set of long-text and short-text benchmark datasets spanning multiple tasks. ChunkLLM not only attains comparable performance on short-text benchmarks but also maintains 98.64% of the performance on long-context benchmarks while preserving a 48.58% key-value cache retention rate. Particularly, ChunkLLM attains a maximum speedup of 4.48x in comparison to the vanilla Transformer in the processing of 120K long texts.

SSDAU: Structured Semantic Data Augmentation for Joint Entity and Relation Extraction

✇cs.AI, q-bio.NC updates on arXiv.org
By: Jiawei He · Mengyu Shi · Jiawei Liu · Zhijie Wang · Chunrong Fang · Xikai Yang · Zhenyu Chen
26 May 2026 at 12:00
arXiv:2605.23440v2 Announce Type: replace-cross Abstract: Joint Entity and Relation Extraction (JERE) is highly susceptible to weak generalization due to low-quality training data. Data augmentation is a common strategy to enhance model generalization across different domains. However, existing data augmentation methods often overlook text relevance and may disrupt semantic structures and dependencies, making it difficult to generate effective augmented data for improving model generalization. In this paper, we propose Structured Semantic Data Augmentation (SSDAU), a novel method designed to preserve the semantic structure of text during augmentation. SSDAU segments text based on entity labels and employs an encoder to capture semantic features of entities through context awareness. It then performs entity semantic restructuring to generate augmented data. To distinguish semantically similar entities, SSDAU fuses contextualized embeddings with traditional similarity scores. To mitigate potential topic ambiguity and information loss, we apply the BERTTopic model to filter out irrelevant topics, ensuring topic consistency. We evaluate SSDAU on datasets with different annotation types and compare its performance on five representative JERE models against seven popular data augmentation baselines. Experiments demonstrate that SSDAU generates semantically consistent data with superior robustness against ambiguity (8.26% F1 decrease vs. 31.91% for baselines), significantly outperforming all existing methods across all metrics.

circPARPBP promotes cancer stemness and chemoresistance in triple-negative breast cancer through recruiting SRCAP complex to activate CCL20 transcription

✇Oncogene - Issue - nature.com science feeds
By: Bo Chen · Kang Ma · Wenhuan Zhong · Yunjie Wang · Cheng Long · Xueyan Mao · Hailin Tang · Xudong Zhu · Anli Yang · Peng Liu
21 May 2026 at 08:00

Oncogene, Published online: 21 May 2026; doi:10.1038/s41388-026-03819-4

circPARPBP promotes cancer stemness and chemoresistance in triple-negative breast cancer through recruiting SRCAP complex to activate CCL20 transcription

Integrative multi-omics and experimental validation reveal UBE2C as a central hub gene and prognostic biomarker in hepatocellular carcinoma

✇Omics in Hepatocellular
By: Lihuang Su · Ouyang Zhang · Yujie Wang · Jiaojian Lv · Qingxiu Liu · Mengchen Ma · Liang Xu · Tongtong Pan
19 May 2026 at 18:00

Int Immunopharmacol. 2026 May 19;183:116866. doi: 10.1016/j.intimp.2026.116866. Online ahead of print.

ABSTRACT

Hepatocellular carcinoma (HCC) is a lethal malignancy with a high recurrence rate and limited treatment options. Ubiquitin-conjugating enzyme E2 C (UBE2C) is implicated in various cancers, yet its impact on the HCC immune landscape remains incompletely understood. Herein, hub genes in HCC were identified, by integrating co-expression networks and protein-protein interaction analyses, from the TCGA, GEO, and CPTAC databases. Their expression was analysed using a single-cell transcriptomic database and verified in HCC tissues and cell lines via quantitative reverse transcription-PCR and immunoblotting. Functional roles of UBE2C were assessed using in vitro knockdown experiments and an in vivo subcutaneous tumour model. The tumour immune microenvironment was profiled using spatial transcriptomics, RNA-seq data, and ssGSEA. A prognostic nomogram was constructed based on multivariate Cox regression. UBE2C was identified as a significantly upregulated hub gene in HCC. Single-cell RNA-seq revealed predominant expression of UBE2C in hepatocytes, with dynamic upregulation along differentiation trajectories. UBE2C knockdown suppressed proliferation, induced apoptosis, and inhibited tumour growth. Spatial transcriptomics highlighted UBE2C-high regions within proliferative niches exhibiting immunosuppressive traits-including TGFB1 enrichment, impaired CXCL9-CXCR3 signalling, and exclusion of cytotoxic T cells-which were reduced in immunotherapy responders. UBE2C expression correlated with immune checkpoint genes and specific immune cell subsets. A UBE2C-based nomogram integrating T stage and tumour stage robustly predicted patient survival, and miR-300 and miR-381-3p were identified as potential upstream regulators. These findings establish UBE2C as a key driver of HCC progression and a biomarker for prognosis and immunotherapy stratification.

PMID:42155390 | DOI:10.1016/j.intimp.2026.116866

Received — 13 April 2026 ⏭

Clinical application of base editing for treating β-thalassaemia

✇Nature - Issue - nature.com science feeds
By: Yongrong Lai · Rongrong Liu · Lijie Wang · Xu-Kai Ma · Yaliang Li · Gaohui Yang · Lingling Shi · Yi-Lin Guo · Zhenbin Wei · Xuemei Zhou · Wenchao Xu · Yaofeng Hou · Annarita Miccio · Bei Yang · Xiaodun Mou · Li Yang · Jia Chen
8 April 2026 at 08:00

Nature, Published online: 08 April 2026; doi:10.1038/s41586-026-10342-9

A clinical phase 1 trial of a single infusion of CS-101, CD34+ cells modified using a transformer base editor to reactivate fetal haemoglobin production, led to early and enduring transfusion independence in patients with β-thalassaemia.

Asymmetric selection of a rice immune module and rebuild of disease resistance

✇Nature - Issue - nature.com science feeds
By: Hui Lin · Fudan Chen · Guanyun Cheng · Bingxiao Yan · Meng Yuan · Jie Qiu · Yiduo Lu · Mingzhe Suo · Ying Chen · Yijie Wang · Kaixuan Cui · Xiangyu Gong · Shasha Liu · Bofan Liu · Jiyun Liu · Jianjun Wang · Rongbai Li · Bizeng Mao · Jianlong Xu · Jong-Seong Jeon · Xuehui Huang · Bin Han · Dong-Lei Yang · Qifei Gao · Haiming Xu · Yiwen Deng · Gongyou Chen · Zuhua He
8 April 2026 at 08:00

Nature, Published online: 08 April 2026; doi:10.1038/s41586-026-10361-6

Stacking XA48-mediated effector-triggered immunity with XA21-mediated pattern-triggered immunity in Oryza sativa japonica reconstitutes the broad-spectrum resistance from wild rice.
Received — 8 April 2026 ⏭

Agile Deliberation: Concept Deliberation for Subjective Visual Classification

✇cs.AI, q-bio.NC updates on arXiv.org
By: Leijie Wang · Otilia Stretcu · Wei Qiao · Thomas Denby · Krishnamurthy Viswanathan · Enming Luo · Chun-Ta Lu · Tushar Dogra · Ranjay Krishna · Ariel Fuxman
7 April 2026 at 12:00
arXiv:2512.10821v2 Announce Type: replace Abstract: From content moderation to content curation, applications requiring vision classifiers for visual concepts are rapidly expanding. Existing human-in-the-loop approaches typically assume users begin with a clear, stable concept understanding to be able to provide high-quality supervision. In reality, users often start with a vague idea and must iteratively refine it through "concept deliberation", a practice we uncovered through structured interviews with content moderation experts. We operationalize the common strategies in deliberation used by real content moderators into a human-in-the-loop framework called "Agile Deliberation" that explicitly supports evolving and subjective concepts. The system supports users in defining the concept for themselves by exposing them to borderline cases. The system does this with two deliberation stages: (1) concept scoping, which decomposes the initial concept into a structured hierarchy of sub-concepts, and (2) concept iteration, which surfaces semantically borderline examples for user reflection and feedback to iteratively align an image classifier with the user's evolving intent. Since concept deliberation is inherently subjective and interactive, we painstakingly evaluate the framework through 18 user sessions, each 1.5h long, rather than standard benchmarking datasets. We find that Agile Deliberation achieves 7.5% higher F1 scores than automated decomposition baselines and more than 3% higher than manual deliberation, while participants reported clearer conceptual understanding and lower cognitive effort.
Received — 4 April 2026 ⏭

Efficient Reasoning with Balanced Thinking

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yulin Li · Tengyao Tu · Li Ding · Junjie Wang · Huiling Zhen · Yixin Chen · Yong Li · Zhuotao Tian
3 April 2026 at 12:00
arXiv:2603.12372v3 Announce Type: replace Abstract: Large Reasoning Models (LRMs) have shown remarkable reasoning capabilities, yet they often suffer from overthinking, expending redundant computational steps on simple problems, or underthinking, failing to explore sufficient reasoning paths despite inherent capabilities. These issues lead to inefficiencies and potential inaccuracies, limiting practical deployment in resource-constrained settings. Existing methods to mitigate overthinking, such as suppressing reflective keywords or adjusting reasoning length, may inadvertently induce underthinking, compromising accuracy. Therefore, we propose ReBalance, a training-free framework that achieves efficient reasoning with balanced thinking. ReBalance leverages confidence as a continuous indicator of reasoning dynamics, identifying overthinking through high confidence variance and underthinking via consistent overconfidence. By aggregating hidden states from a small-scale dataset into reasoning mode prototypes, we compute a steering vector to guide LRMs' reasoning trajectories. A dynamic control function modulates this vector's strength and direction based on real-time confidence, pruning redundancy during overthinking, and promoting exploration during underthinking. Extensive experiments conducted on four models ranging from 0.5B to 32B, and across nine benchmarks in math reasoning, general question answering, and coding tasks demonstrate that ReBalance effectively reduces output redundancy while improving accuracy, offering a general, training-free, and plug-and-play strategy for efficient and robust LRM deployment. Project page and code are available at https://rebalance-ai.github.io .
❌