Academic Intelligence · Curated Daily

Explore the Frontier of Global Academia

AcademicHub aggregates real-time literature from top journals and preprint platforms. Build your personal research radar and let large language models compile cross-disciplinary analysis briefings automatically.

01.
bioRxiv (Bioinfo) 2026-06-21

Expanding the GUSome: Structure-guided identification and characterization of gut microbial β-glucuronidases

The gut microbiome-encoded {beta}-glucuronidase (GUS) enzymes have a significant effect on human physiology through their deglucuronidation activity on endogenous and exogenous glucuronides. GUS activity also significantly influences the pharmacokinetics, efficacy and toxicity of various drugs including chemotherapeutic drugs. Given their crucial role in drug metabolism, GUS enzymes have emerged as promising targets for therapeutic intervention. Here, we have identified and characterized 79 unique GUS enzymes through a structure-guided approach. Structural modelling of these GUS enzymes revealed a conserved core and active-site residues with significant variations in the number and nature of the C-terminal domains. A new classification system based on the number and type of additional C-terminal domains is presented for the GUS proteins. Further, GUS enzymes have been categorized into different loop categories linked to their substrate preferences. The relationship between domain architecture and loop-type is explored by sequence similarity network analysis. We could successfully express, purify and validate GUS processing capability of a panel of identified GUS proteins. The nature of oligomer organization has been deciphered by SEC and DLS studies. Further, we have identified additional GUS enzymes capable of processing SN-38G, glucuronidated form of anticancer drug, irinotecan. These newly identified GUS enzymes will offer valuable insights into gut microbial GUS diversity and their role in understanding the population-specific drug-induced adverse effects on human health.

02.
arXiv (CS.CL) 2026-06-12

Multi-Turn Reasoning When Context Arrives in Pieces: Scalable Sharding and Memory-Augmented RL

When a user reveals task-critical information across several conversation turns, LLM accuracy drops by up to 65% despite full context availability. We show that this Lost in Conversation degradation can be substantially mitigated by training models to maintain a compact rolling memory instead of attending to a growing history. To make such training scalable, we introduce a low-cost sharding pipeline that converts single-turn QA datasets into multi-turn fragmented-information episodes, eliminating the need for hours of manual annotation. Training only on sharded GSM8K, our memory-augmented policy significantly improves multi-turn accuracy and generalises zero-shot to harder math and out-of-domain long-context QA. Moreover, memory-trained models outperform full-history baselines even when given the full history at test time, suggesting that learning to compress induces more robust incremental reasoning than full-context exposure alone.

03.
arXiv (CS.CL) 2026-06-11

Fanar-Sadiq: A Multi-Agent Architecture for Grounded Islamic QA

Large language models (LLMs) can answer religious knowledge queries fluently, yet they often hallucinate and misattribute sources, which is especially consequential in Islamic settings where users expect grounding in canonical texts (Qur'an and Hadith) and jurisprudential (fiqh) nuance. Retrieval-augmented generation (RAG) improves grounding, however, a single retrieve-then-generate pipeline is insufficient for diverse Islamic queries, including verbatim scripture, citation-grounded guidance, and rule-constrained computations such as zakat and inheritance. To address these challenges, we present Fanar-Sadiq, a bilingual Arabic-English Islamic QA system built on a multi-agent, tool-augmented architecture. It is a core component of the Fanar AI platform. Fanar-Sadiq routes Islamic queries to specialized modules within an agentic tool architecture. It supports intent-aware routing, retrieval-grounded fiqh answers with normalized citations and verification traces, exact verse lookup with quotation validation, and deterministic Sunni zakat and inheritance calculators with madhhab-sensitive branching. We evaluate the end-to-end system on public Islamic QA benchmarks and show strong effectiveness and efficiency. It is publicly accessible through an API and Web application and has received over 1.9M accesses in less than a year (https://api.fanar.qa/docs).

04.
medRxiv (Medicine) 2026-06-24

Barriers and facilitators to diabetes management among adults and healthcare providers in a peri-urban Ugandan health facility: A qualitative study

Diabetes mellitus is an increasing public health challenge in Uganda and other low- and middle-income countries, where health systems face growing demands for chronic disease care. Although quantitative studies have documented poor glycemic control and health system constraints, less is known about how patients and healthcare providers experience diabetes management in peri-urban public health settings. This study explored barriers and facilitators to diabetes management among adults with diabetes mellitus and healthcare providers at a peri-urban health facility in Uganda. We conducted a qualitative descriptive study at Kasangati Health Centre IV, Wakiso District, Uganda, between February and March 2025. Data were collected through 15 in-depth interviews with adults living with diabetes mellitus and 8 key informant interviews with healthcare providers involved in diabetes care. Participants were purposively selected based on their experience with diabetes management and service delivery. Interviews were audio-recorded, transcribed verbatim, translated where necessary, and analyzed using a hybrid inductive-deductive thematic approach informed by the Theoretical Domains Framework. Five interrelated themes were identified: (1) institutional and environmental factors influencing access to diabetes care; (2) cognitive and informational factors influencing medication adherence; (3) social influences on diabetes management; (4) emotional experiences of patients and healthcare providers; and (5) self-management strategies and continuity of care. Across these themes, participants identified barriers including resource limitations, communication challenges, medication management difficulties, stigma, emotional distress, and weak follow-up systems. Facilitators included peer support, religious and community networks, health education, provider flexibility, and patient-developed adherence strategies. Diabetes management was influenced by interacting health-system, social, informational, and behavioural factors. Resource constraints, limited health literacy, stigma, and weak follow-up systems hindered effective management, while social support, health education, and patient self-management strategies facilitated continued engagement in care. Interventions that strengthen chronic care services, patient education, and community support may improve diabetes outcomes in similar resource-constrained settings.

05.
arXiv (CS.CV) 2026-06-18

Spatially Stratified Distillation for Heterogeneous Radar Place Recognition

Scalable, all-weather place recognition increasingly relies on heterogeneous radar place recognition to bridge diverse hardware platforms. A notable application is matching queries from cost-effective 4D automotive radars against high-fidelity reference maps built by dense spinning radars. This process is fundamentally limited by the extreme sparsity (and narrow field-of-view) of the 4D sensor, which captures only a fraction of the structural density present in the spinning radar database. Prior efforts address this issue by unifying different radar signals. That is, projecting both signals into a common representational space. Yet, they suffer performance degradation in multi-session environments. In this paper, we propose spatially-stratified distillation (SSD); a strategy that replaces standard uniform distillation with an asymmetric spatial alignment derived directly from physical radar returns. In regions where both radars exhibit overlapping returns, SSD enforces strong feature alignment. Crucially, in sparse regions where the 4D student lacks returns but the teacher contains valid structure within the shared field of view, SSD applies heavily discounted distillation weights. Extensive evaluations of the recent HeRCULES dataset demonstrate that SSD significantly outperforms prior place recognition methods, achieving state-of-the-art results on its challenging dynamic sequences.

06.
arXiv (CS.AI) 2026-06-12

Proprioceptive-visual correspondence enables self-other distinction in humanoid robots

arXiv:2606.13222v1 Announce Type: cross Abstract: Distinguishing self from others is a prerequisite for social intelligence, yet humanoid robots that increasingly share workspaces with humans still lack this ability. Here we show that a humanoid robot can learn self-other distinction from proprioceptive-visual correspondence, without any identity labels or kinematic models. Once established, this distinction bootstraps a predictive self-model that maps joint configurations to three-dimensional body occupancy, capturing how the robot's body changes with action. In multi-agent scenes involving humans or morphologically identical robots, the system reliably identifies itself, learns a 3D self-model, and supports downstream tasks including target reaching, collision-aware motion planning, and human-to-robot motion retargeting. Together, these results outline a route toward bodily self-representation in robots that act and coordinate alongside others in shared physical environments. Project page: https://euron-zc.github.io/humanoid-self-model/.

07.
arXiv (CS.CL) 2026-06-25

OPERA: Aligning Open-Ended Reasoning via Objective Perplexity-based Reinforcement Learning

Reinforcement Learning (RL) has enabled LLMs to excel in objective reasoning tasks such as mathematics and code generation. However, applying RL to open-ended tasks, such as creative writing, remains challenging because LLM-as-a-judge reward models often exhibit stylistic biases and positional inconsistencies, leading to unstable supervision. To address this, we propose OPERA (Objective Perplexity-based Reflective Alignment), which replaces unreliable external judges with intrinsic rewards derived from perplexity dynamics. Specifically, we derive an intrinsic reward signal from perplexity dynamics, quantifying uncertainty reduction at critical reflective states. During the cold-start phase, we introduce a data synthesis method that leverages carefully designed guiding words to generate diverse reasoning traces, along with perplexity-prioritized rollouts that utilize internal log-probabilities to identify logically consistent reasoning branches. This pipeline yields a large-scale dataset comprising 20,000 high-quality reasoning trajectories. Empirical evaluations consistently demonstrate the scalability and efficacy of our approach in alignment for open-ended tasks. Implementing OPERA on Qwen3-8B establishes a new state-of-the-art among open-source models, achieving parity with or surpassing proprietary models like Gemini2.5 and MiniMax-M2.5 in some open-ended tasks. The code is available at https://github.com/pangpang-xuan/OPERA.

08.
arXiv (CS.AI) 2026-06-16

Towards Verifiable Agentic Data Science: Solving Irregular TSQA Via Tool-Grounded Reasoning

arXiv:2606.15107v1 Announce Type: new Abstract: Time series data in real-world deployments is overwhelmingly irregular. Observations are asynchronous, missing values are informative rather than random, and sampling frequencies vary across sensors and operational windows. However, existing Time Series Question Answering (TSQA) benchmarks mostly assume regularly sampled inputs, leaving a fundamental gap in understanding how large language models (LLMs) and AI agents perform under irregular conditions. To bridge this gap, we introduce IRTS-ToolBench, a benchmark of 1,700 questions spanning 10 task types across 13 domains. IRTS-ToolBench is designed to be used independently by any researcher working on LLM-based irregular time series analysis, providing standardized inputs and a reproducible evaluation protocol. Code can be found in https://github.com/SanhornC/IRTS-ToolBench.

09.
arXiv (CS.CV) 2026-06-25

Spatio-Temporal Mixture-of-Modality-Experts Diffusion for Quantitative DCE-MRI Synthesis from Incomplete MR Sequences

Quantitative maps from dynamic contrast-enhanced MRI (DCE-MRI) are essential for tumor assessment but are often unavailable due to contrast-agent risks and protocol variability. Prior methods predict these maps from other MRI modalities, yet most assume fixed, fully observed inputs and fail under realistic missingness. We present Spatio-Temporal Mixture-of-Modality-Experts (ST-MoME), a conditional diffusion framework that synthesizes 3D DCE parameter maps from diverse subsets of multimodal MRI. ST-MoME fuses modality-specific expert features through a spatio-temporal gating network that produces voxel-wise, timestep-dependent weights, forming a conditioning tensor that guides denoising. To preserve quantitative fidelity, ST-MoME performs diffusion directly in image space with 3D patch-based training and a Swin-based backbone. On a clinical brain-tumor cohort of 386 patients, we evaluate ST-MoME across 16 controlled modality-availability scenarios. It achieves the lowest mean Normalized Mean Square Error (NMSE) aggregated across all three DCE parameters, with leading performance on $v_p$ and $v_e$, competitive results on $K^{\mathrm{trans}}$, and the lowest reconstruction error within the clinically critical tumor region. A post-hoc analysis of the learned gating dynamics shows a structural-early, physiological-late fusion schedule consistent with clinical intuition.

10.
arXiv (CS.CL) 2026-06-12

SkillChain: Closing the Loop on Skill Evolution for Image-Based E-Commerce AI Assistants

Image-based AI assistants are now deployed at production scale on e-commerce platforms, where a single uploaded image can trigger fundamentally different user intents: product search, style recommendation, visual encyclopedia, or utility tool calls, each demanding its own response format, tool invocation, and domain knowledge. Without per-intent behavioral constraints, LLM-based systems conflate these heterogeneous modes and fall short of domain quality standards, while the breadth and dynamism of the intent space render manual engineering infeasible. To address this, we present SkillChain, which closes the production feedback loop on Skill evolution, automating the lifecycle of Skills through three stages: Skill Creator for bootstrapping from task specs and trajectories, Route Optimizer for routing alignment, and Body Refiner for iterative Skill Body refinement via dual-path LLM-Judge evaluation. Deployed on a production-scale e-commerce image assistant, SkillChain substantially improves aggregate response quality, with the strongest gains on structural compliance and content quality; a one-week online A/B experiment further confirms significant gains in user engagement, content consumption, and long-term retention.

11.
arXiv (quant-ph) 2026-06-25

Rapid and robust laser-frequency auto-locking using Bayesian-optimization and discrete-wavelet-transformation algorithms

arXiv:2606.25267v1 Announce Type: new Abstract: Rapid and robust laser-frequency auto-locking is essential for the field deployment of quantum communications, quantum computing, and precision-measurement technologies; however, achieving this remains a considerable challenge. Here, we propose and demonstrate an auto-locking scheme employing Bayesian optimization and discrete biorthogonal wavelet transformation. First, the reference is rapidly sought by making intelligent use of historical observations, eliminating the inherent blindness of the traditional parameter-scanning method. Second, the frequency reference is robustly identified by pinpointing transition signals with the discrete biorthogonal wavelet transformation and analyzing their immutable frequency differences and relative magnitudes, which are determined by the inherent atomic structure and remain resistant to environmental disturbances. This proposed approach achieves a fivefold acceleration in reference searching compared to conventional scanning methods in the case where the laser frequency drifts far away from the reference. Crucially, it achieves an identification accuracy of more than 99.5 %, even under severe 50 % laser-intensity fluctuations, $9.95^\circ$ photodiode misalignment, and $18^\circ$C Rb cell temperature elevation. Finally, locking the laser frequency to the identified reference with a lead zirconate titanate-current double-servo loop narrows the linewidth to 20 kHz. We believe that this rapid, robust, and high-performance auto-locking technique will be pivotal towards the deployment of the next generation of practical quantum technologies in demanding field environments.

13.
arXiv (CS.AI) 2026-06-16

RaBiT: Residual-Aware Binarization Training for Accurate and Efficient LLMs

arXiv:2602.05367v3 Announce Type: replace Abstract: Efficient deployment of large language models (LLMs) requires extreme quantization, forcing a critical trade-off between low-bit efficiency and performance. Residual binarization enables hardware-friendly, matmul-free inference by stacking binary ($\pm$1) layers, but is plagued by pathological feature co-adaptation. We identify a key failure mode, which we term inter-path adaptation: during quantization-aware training (QAT), parallel residual binary paths learn redundant features, degrading the error-compensation structure and limiting the expressive capacity of the model. While prior work relies on heuristic workarounds (e.g., path freezing) that constrain the solution space, we propose RaBiT, a novel quantization framework that resolves co-adaptation by algorithmically enforcing a residual hierarchy. Its core mechanism sequentially derives each binary path from a single shared full-precision weight, which ensures that every path corrects the error of the preceding one. This process is stabilized by a robust initialization that prioritizes functional preservation over mere weight approximation. RaBiT redefines the 2-bit accuracy-efficiency frontier: it achieves state-of-the-art performance, rivals even hardware-intensive Vector Quantization (VQ) methods, and delivers a $4.49\times$ inference speed-up over full-precision models on an RTX 4090. Code is available at https://github.com/SamsungLabs/RaBiT.

14.
arXiv (CS.CL) 2026-06-16

Scaling Human and G2P Supervision for Robust Phonetic Transcription

Expert phonetic annotation is costly, especially for non-standard dialects and atypical speech. A common alternative is using Grapheme-to-Phoneme (G2P) models to auto-generate phonetic labels from text transcripts at scale. We study how automatic phonetic transcription performance scales with human and G2P supervision in English. Using a curated 80-hour benchmark spanning native, non-native and post-stroke speech, we identify a supervision quality threshold: G2P supervision helps only when fewer than 20-30 hours of human annotation are available. Beyond this threshold, it provides no significant benefit and can reduce cross-dialect robustness. What is effective after this threshold is ASR pretraining which we use to achieve a 2.3x reduction in weighted phone feature error rate over prior systems, with strong gains on non-native and aphasic speech. These results suggest that quantity-driven G2P scaling may yield diminishing returns for robust generalization.

15.
arXiv (quant-ph) 2026-06-16

Hyperinvariant Spin Network States – An AdS/CFT Model from First Principles

arXiv:2510.06602v2 Announce Type: replace Abstract: We study the existence and limitations of hyperinvariant tensor networks incorporating a local SU(2) symmetry. As discrete implementations of the anti de-Sitter/conformal field theory (AdS/CFT) correspondence, such networks have created bridges between the fields of quantum information theory and quantum gravity. Adding SU(2) symmetry to the tensor network allows a direct connection to spin network states, a basis of the kinematic Hilbert space of loop quantum gravity (LQG). We consider a particular situation where the states can be interpreted as kinematic quantum states for three-dimensional quantum gravity. We show that important aspects of the AdS/CFT correspondence are realized in certain quantum states of the gravitational field in LQG, thus justifying, from first principles, a class of models introduced by [F. Pastawski et al., JHEP 06, 149 (2015)]. We provide examples of hyperinvariant tensor networks, but also prove constraints on their existence in the form of no-go theorems that exclude absolutely maximally entangled states as well as general holographic codes from local SU(2)-invariance. We calculate surface areas as expectation values of the LQG area operator and discuss further possible constraints as a consequence of a decay of correlations on the boundary.

16.
arXiv (CS.AI) 2026-06-16

Continuous Cross-Domain Traffic State Prediction via Memory-Augmented Graph Liquid Time-Constant Networks

arXiv:2606.15807v1 Announce Type: cross Abstract: Traffic state prediction is a fundamental task in intelligent transportation systems. In practical applications, some regions suffer from limited traffic observations due to insufficient sensing infrastructure, making cross-domain knowledge transfer an important solution for data-scarce traffic prediction. However, existing cross-domain traffic prediction methods still face several limitations, including coarse-grained source-target adaptation, limited capability in handling unseen target-domain patterns, and insufficient modeling of continuous traffic dynamics under irregular or heterogeneous temporal conditions. To address these issues, this paper proposes a continuous cross-domain traffic prediction framework, termed Memory-Augmented Graph Liquid Time-Constant Network (MA-GLTC). Specifically, we first construct spatio-temporal units (STUs) to decompose traffic networks into transferable local units, enabling fine-grained knowledge alignment across domains. Then, a graph liquid time-constant network (GLTC) is developed to model graph-coupled traffic evolution in continuous time. Different from generic graph neural ODE-based models, GLTC introduces graph-coupled recurrent conductance into liquid time-constant dynamics, allowing node states to evolve with leakage, adaptive time constants, and neighborhood-aware feedback. Furthermore, a Memory-based Transfer Storage (MTS) mechanism is designed to preserve source-domain knowledge, retrieve matched traffic patterns, and update reliable target-domain patterns when unseen states emerge. Experiments on five public traffic datasets demonstrate that MA-GLTC consistently outperforms representative innerdomain and cross-domain baselines in both short-term and longterm prediction tasks. Compared with the second-best method, MA-GLTC reduces the average prediction errors by 3.02%, 0.33%, 8.92%, 10.09%, and 2.11%, respectively.

17.
arXiv (CS.CL) 2026-06-25

A Systematic Analysis of Hybrid Linear Attention

Transformers face quadratic complexity and memory issues with long sequences, prompting the adoption of linear attention mechanisms using fixed-size hidden states. However, linear models often suffer from limited recall performance, leading to hybrid architectures that combine linear and full attention layers. Despite extensive hybrid architecture research, the choice of linear attention component has not been deeply explored. We systematically evaluate various linear attention models across generations - vector recurrences to advanced gating mechanisms - both standalone and hybridized. To enable this comprehensive analysis, we trained and open-sourced 72 models: 36 at 340M parameters (20B tokens) and 36 at 1.3B parameters (100B tokens), covering six linear attention variants across five hybridization ratios. Benchmarking on standard language modeling and recall tasks reveals that superior standalone linear models do not necessarily excel in hybrids. While language modeling remains stable across linear-to-full attention ratios, recall significantly improves with increased full attention layers, particularly below a 3:1 ratio. Our study highlights selective gating, hierarchical recurrence, and controlled forgetting as critical for effective hybrid models. We recommend architectures such as HGRN-2 or GatedDeltaNet with a linear-to-full ratio between 3:1 and 6:1 to achieve Transformer-level recall efficiently. Our models are open-sourced at https://huggingface.co/collections/m-a-p/hybrid-linear-attention-research-686c488a63d609d2f20e2b1e.

18.
arXiv (CS.LG) 2026-06-18

CODEBLOCK: Learning to Supervise Code at the Right Granularity

arXiv:2606.18286v1 Announce Type: new Abstract: Supervised fine-tuning of code LLMs typically applies uniform cross-entropy loss to all response tokens, implicitly assuming that every token provides equally useful learning signal. Recent token-level selection methods challenge this assumption in natural-language SFT by supervising only high-value tokens. However, directly transferring token-level masking to code can break syntactically and semantically coherent program units, because code depends on structural completeness and definition-use relations. We therefore propose CodeBlock, a structure-aware sparse supervision framework that selects structure-complete code evidence rather than isolated tokens. CodeBlock first selects high-quality instruction-response pairs, then partitions code responses into syntactically coherent coding items, estimates their utility by aggregating generalized cross-entropy over core logic tokens, and reranks them with data-flow reach and bridge signals to prioritize blocks that propagate or connect important program dependencies. During training, the full response remains available as context, while loss is applied only to selected code items and informative natural-language tokens. Experiments on six code-generation benchmarks show that CodeBlock achieves stronger average pass@1 than full-token SFT and competitive selection baselines, while using only 1.9% of supervised response tokens.

19.
arXiv (CS.LG) 2026-06-18

Latent-Conditioned Parameterized Quantum Circuits as Universal Approximators for Distributions over Quantum States

arXiv:2605.28690v3 Announce Type: replace-cross Abstract: Many applications in quantum simulation, quantum chemistry, and quantum machine learning require not a single quantum state but an ensemble of states characterizing the heterogeneity of a target system. Preparing such ensembles state-by-state is prohibitive in both variational and fault-tolerant settings, thereby motivating a generative modeling approach. We introduce latent-conditioned parameterized quantum circuits (LPQCs), a hybrid quantum-classical framework in which classical neural networks map a latent variable sampled from a prior distribution to the parameters of a parameterized quantum circuit. We prove that LPQCs are universal approximators for probability measures over density operators in the 1-Wasserstein distance, extending classical universal approximation theorems to the quantum-distribution setting. We additionally introduce a multimodal latent prior and a mixture-of-experts circuit architecture, and show empirically that the latent-conditioned parameterization alleviates the barren plateau problem during optimization, a behavior for which we provide rigorous partial guarantees. Numerical experiments validate the framework on a synthetic multi-cluster ensemble of mixed quantum states and on a QM9-derived ensemble of 3-D molecular structures. In these tasks, LPQC outperforms recent quantum generative baselines and matches the generation quality of a classical neural-network baseline, while requiring an output dimension that grows only linearly with the number of qubits rather than exponentially. By leveraging classical expressivity in the latent space, LPQCs offer a tractable route to quantum generative modeling.

20.
arXiv (CS.CV) 2026-06-15

Clay-CNN Hybrids: Leveraging Geo-Foundational Models as Auxiliary Context for Landslide Detection

Authors:

Rapid post-event landslide mapping is essential for disaster response but remains difficult to automate due to extreme class imbalance. This study evaluates whether Clay v1.5, a Geo-Foundational Model (GFM), can improve pixel-level landslide segmentation on the Landslide4Sense (L4S) benchmark, which contains 3,799 training chips with 14 Sentinel-2 and terrain bands and approximately 2% positive pixels. We compare three strategies: Clay as the primary encoder with multi-scale residual terrain fusion, a U-Net backbone augmented with Clay semantic context at the bottleneck, and a standard U-Net baseline. The hybrid U-Net + Clay model with two-stage Low-Rank Adaptation (LoRA) achieved the best test F1 of 64.5 +/- 1.8% over three seeds, surpassing the Clay-only backbone (55.2 +/- 3.6%) and the U-Net baseline (59.9%). Clay as a standalone encoder underperformed the U-Net due to the absence of multi-scale skip connections, but its pretrained representations consistently improved performance when injected as auxiliary context. These findings suggest that GFMs are most effective for landslide detection when they complement spatially detailed convolutional architectures rather than replace them.

21.
arXiv (CS.LG) 2026-06-19

Approximate Next Policy Sampling: Replacing Conservative Target Policy Updates in Deep RL

arXiv:2605.05481v2 Announce Type: replace Abstract: We revisit a classic "chicken-and-egg" problem in reinforcement learning: to safely improve a policy, the value function must be accurate on the state-visitation distribution of the updated policy. That distribution over states is unknown and cannot be sampled for the purposes of training the value function. Conservative updates solve this problem, but at the cost of shrinking the policy update. This paper explores an alternative solution, Approximate Next Policy Sampling (ANPS), which addresses the problem by modifying the training distribution rather than constraining the policy update. ANPS is satisfied if the distribution of the training data approximates that of the next policy. To demonstrate the feasibility and efficacy of ANPS, we introduce Stable Value Approximate Policy Iteration (SV-API). SV-API modifies the standard approximate policy iteration loop to hold the target policy fixed while an iteratively updated behavioral policy gathers relevant experience. It only commits to a new policy once a convergence criterion has been met. If certain stability criteria are met, the update is guaranteed to be safe; otherwise, it remains no less safe than standard approximate policy iteration. Applying SV-API to PPO yields Stable Value PPO (SV-PPO), which matches or improves performance on high-dimensional discrete (Atari) and continuous control benchmarks while executing substantially larger target policy updates. These results demonstrate the viability of ANPS as a new solution to this classic challenge in RL.

22.
arXiv (CS.AI) 2026-06-19

MetaResearcher: Scaling Deep Research via Self-Reflective Reinforcement Learning in Adversarial Virtual Environments

arXiv:2606.19893v1 Announce Type: new Abstract: Deep research agents have demonstrated remarkable capabilities in autonomous information gathering and synthesis, yet their training remains constrained by the static nature of simulated environments, the limits of fact-retrieval-only task designs, and the inefficiency of outcome-based reinforcement learning. In this work, we propose MetaResearcher, a novel framework that scales deep research agent training across four synergistic dimensions. First, we introduce an Evolving Virtual World that injects temporal dynamics and adversarial misinformation into the training environment, forcing agents to develop source credibility assessment and temporal conflict resolution skills. Second, we design Discovery-Oriented Tasks – including hypothesis generation and contradiction resolution – that transcend simple fact retrieval and push agents toward genuine research behaviors. Third, we propose a Self-Reflective Meta-Reward mechanism within the GRPO framework that jointly optimizes for answer correctness, search path efficiency, reflection depth, and tool call diversity, directly addressing the repetitive action loop problem observed in prior work. Fourth, we introduce a Heterogeneous Multi-Agent Swarm architecture comprising specialized Scout, Filter, and Synthesizer models that learn collaborative research strategies through coordinated reinforcement learning. Built upon the LiteResearcher infrastructure, MetaResearcher requires zero marginal API cost for training while targeting substantial improvements in both benchmark performance (GAIA, Xbench-DS) and epistemic robustness under adversarial conditions. We present the complete framework design, training methodology, and planned experimental validation.

23.
arXiv (CS.CV) 2026-06-25

Entropy-Based Observability for AI Agent Behavior

AI agents are typically instrumented through outcome-oriented indicators such as task success, reward, latency, and cost.Although these indicators are operationally important, they provide limited visibility into the internal structure of agent behavior such as the degree of exploration, the rigidity or diversity of action selection, the concentration of tool use, the reduction of uncertainty across a run, and the stability of behavior across repeated executions.This paper proposes Entropy-Based Observability for AI Agents (EOA), a lightweight framework for deriving behavioral telemetry from agent traces.

24.
arXiv (CS.AI) 2026-06-25

Bias Fitting to Mitigate Length Bias of Reward Model in RLHF

arXiv:2505.12843v2 Announce Type: replace-cross Abstract: Reinforcement Learning from Human Feedback (RLHF) relies on reward models to align large language models with human preferences. However, RLHF often suffers from reward hacking, wherein policy learning exploits flaws in the trained reward model to maximize reward scores without genuinely aligning with human preferences. A significant example of such reward hacking is length bias, where reward models usually favor longer responses irrespective of actual response quality. Previous works on tackling length bias have notable limitations, these approaches either mitigate bias without characterizing the bias form, or simply assume a linear length-reward relation. To accurately model the intricate nature of length bias and facilitate more effective bias mitigation, we propose FiMi-RM (Bias Fitting to Mitigate Length Bias of Reward Model), a framework that autonomously learns and corrects underlying bias patterns. Our approach consists of three stages: First, we warm up by training a standard reward model which inherently contains length bias. Next, we deploy a lightweight fitting model to capture the non-linear relation between length and reward. Finally, we incorporate this learned relation into the reward model, effectively decoupling length from reward while preserving preference modeling capabilities. Experimental results demonstrate that FiMi-RM achieves a more balanced length-reward distribution. Furthermore, when applied to alignment algorithms such as Direct Preference Optimization (DPO) and Best-of-N (BoN), our debiased reward model improves length-controlled win rate and reduces verbosity without compromising its performance.

25.
arXiv (CS.CV) 2026-06-16

Post-Launch Capability Expansion of Vision-Language Models via Prompting for On-Orbit Spacecraft Inspection

Spaceborne inspection systems often deploy perception models prior to launch, after which updating model weights or expanding fixed label sets becomes operationally impractical. While supervised models can be integrated pre-flight, adding new semantic capabilities in orbit requires retraining and re-uploading parameters. We investigate whether prompt-driven vision–language models can enable post-launch semantic expansion, allowing new spacecraft components to be specified via natural-language prompts without modifying onboard weights. We evaluate zero-shot instance segmentation of spacecraft components under a strictly frozen, single-pass inference protocol on a test set of $129$ images of previously unseen satellites. Under fixed global thresholds and no post-processing, SAM3 achieves $0.385$ mAP@$0.5$ and $0.267$ mAP@$0.5{:}0.95$. Performance is strongly scale-dependent: large structural elements like spacecraft bodies ($0.639$ AP@$0.50$) and solar arrays ($0.598$ AP@$0.5$) localize reliably, while relatively small appendages like antennas ($0.221$ AP@$0.5$) and thrusters ($0.081$ AP@$0.5$) remain difficult. Prompt formulation influences performance, with structured prompts incorporating spatial and geometric descriptors yielding up to $82%$ improvement over short category-name prompts. The model operates within the memory and compute envelope of contemporary embedded GPUs, suggesting prompt-driven grounding can provide a practical mechanism for post-launch semantic extension of dominant spacecraft structures while highlighting limitations of zero-shot localization for fine-scale components under orbital domain shift.