Academic Intelligence · Curated Daily

探索全球前沿学术脉络

AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达,利用大语言模型自动生成交叉领域文献分析简报。

01.
arXiv (CS.LG) 2026-06-19

AgentArmor: A Framework, Evaluation, \& Mitigation of Coding Agent Failures

arXiv:2606.19380v1 Announce Type: cross Abstract: Software engineering and deployment are increasingly being delegated to AI coding agents. The scale of their adoption is surfacing rare, but highly destructive, failure modes. In this paper, we study these failure modes as stemming from three distinct mechanisms: underspecification, where default model behavior is unsafe; capability errors, where the safe action is available but the model does not adhere to it due to bias or capability limitations; and agent harness errors, where the model fails to execute the safe action through the harness. We evaluate these across 8 different evaluations, each inspired by real-life deployment failures, totaling 20 coding environments and 59 synthetic transcript templates. Based on this evaluation, we propose AgentArmor, an agent harness modification, to mitigate these errors. By adding an extended system prompt, a separate command classifier, a ``3 strikes'' policy, deterministic guardrails, and tools for the agent to edit its own context, we show that AgentArmor is safer across a statistically significant number of samples. Thus, we suggest concrete mitigations for current coding agents and a design philosophy for future agent harness features.

03.
arXiv (CS.CL) 2026-06-24

Preferences of a Voice-First Nation: Large-Scale Pairwise Evaluation and Preference Analysis for TTS in Indian Languages

Crowdsourced pairwise evaluation has emerged as a scalable approach for assessing foundation models. However, applying it to Text to Speech(TTS) introduces high variance due to linguistic diversity and multidimensional nature of speech perception. We present a controlled multidimensional pairwise evaluation framework for multilingual TTS that combines linguistic control with perceptually grounded annotation. Using 5K+ native and code-mixed sentences across 10 Indic languages, we evaluate 7 state-of-the-art TTS systems and collect over 120K pairwise comparisons from over 1900 native raters. In addition to overall preference, raters provide judgments across 6 perceptual dimensions: intelligibility, expressiveness, voice quality, liveliness, noise, and hallucinations. Using Bradley-Terry modeling, we construct a multilingual leaderboard, interpret human preference using SHAP analysis and analyze leaderboard reliability alongside model strengths and trade-offs across perceptual dimensions.

04.
arXiv (CS.CV) 2026-06-18

MUFASA: A Multi-Layer Framework for Slot Attention

Unsupervised object-centric learning (OCL) decomposes visual scenes into distinct entities. Slot attention is a popular approach that represents individual objects as latent vectors, called slots. Current methods obtain these slot representations solely from the last layer of a pre-trained vision transformer (ViT), ignoring valuable, semantically rich information encoded across the other layers. To better utilize this latent semantic information, we introduce MUFASA, a lightweight plug-and-play framework for slot-attention-based approaches to unsupervised object segmentation. Our model computes slot attention across multiple feature layers of the ViT encoder, fully leveraging their semantic richness. We propose a fusion strategy to aggregate slots obtained on multiple layers into a unified object-centric representation. Integrating MUFASA into existing OCL methods improves their segmentation results across multiple datasets, setting a new state of the art while simultaneously improving training convergence with only minor inference overhead.

05.
arXiv (CS.CL) 2026-06-11

I Understand How You Feel: Enhancing Deeper Emotional Support Through Multilingual Emotional Validation in Dialogue System

Emotional validation - explicitly acknowledging that a user's feelings make sense - has proven therapeutic value but has received little computational attention. Emotional validation in dialogue systems can be decomposed into (i) validating response identification, (ii) validation timing detection, and (iii) validating response generation. To support research on all three subtasks, we release M-EDESConv, a 120k English-Japanese multilingual corpus created through hybrid manual and automatic annotation, and M-TESC, a multilingual spoken-dialogue test set. For timing detection, we propose MEGUMI, a Multilingual Emotion-aware Gated Unit for Mutual Integration, that fuses frozen XLM-RoBERTa semantics with language-specific emotion encoders via cross-modal attention and gated fusion. MEGUMI shows superior performance on both the M-EDESConv and M-TESC datasets, both objectively and subjectively. Finally, our EmoValidBench benchmarks of GPT-4.1 Nano and Llama-3.1 8B indicate that current LLMs generate contextually similar and diverse validating responses, but emotional understanding remains a major area for improvement. Project page: https://github.com/zihaurpang/Multilingual-Emotional-Validation

06.
arXiv (CS.LG) 2026-06-24

PROTECT-90: A Fault Dataset for Power System Protection

arXiv:2606.24298v1 Announce Type: cross Abstract: The increasing interest in data-driven methods for power system protection is accompanied by a lack of standardized, publicly available high-voltage waveform datasets that enable transparent and reproducible evaluation. To address this gap, this paper introduces the PROTECT-90 dataset, an open electromagnetic transient (EMT)-simulated reference benchmark for high-voltage fault studies with consistent digital-fault-recorder-like measurements, publicly released with this work. The dataset comprises 9,022 physically consistent short-circuit simulation episodes generated on a standardized 90 kV double-line topology with systematically documented domain randomization of grid operating points, line parameters, and fault conditions. For each episode, synchronized three-phase voltage and current waveforms are recorded at eight measurement locations and released together with structured, machine-readable metadata describing fault type, fault location, inception time, and operating conditions. All modeling assumptions, parameter ranges, and data-generation procedures are explicitly documented to ensure transparency and cross-study comparability. By combining physically grounded EMT simulation, balanced scenario coverage, and open accessibility, PROTECT-90 establishes a standardized foundation for reproducible benchmarking of protection-oriented signal processing and learning-based methods.

07.
arXiv (CS.LG) 2026-06-12

Viral Proteins Reveal Geometry of Protein Language Models

arXiv:2606.12609v1 Announce Type: new Abstract: Protein language models are trained on highly imbalanced datasets, raising the question of how they represent underrepresented biological sequences. Using viral proteins as a case study across ESM model families, we identify a dominant nativeness axis in embedding space, aligned with masked reconstruction perplexity, that orders sequences from well-modeled cellular proteins through viral proteins to shuffled and random sequences. Scaling contracts this axis unevenly across viral families. Despite this, protein language model embeddings retain viral-specific signal: viral proteins remain linearly separable beyond zero-shot perplexity and shallow sequence features. Together, these results suggest that pLM representations are structured by a general notion of nativeness while preserving information specific to distinct biological groups.

08.
medRxiv (Medicine) 2026-06-23

Attention and memory in Parkinson's disease: a discriminant analysis approach

Background. Cognitive impairment in Parkinson's disease (PD) is highly prevalent and heterogeneous. Assessing multiple cognitive domains is challenging and risks redundancy. This study evaluated whether a discriminant analysis approach could optimize the selection of specific tasks and measures for identifying attention and memory deficits in PD. Methods. Thirty PD patients and 25 cognitively unimpaired (CU) controls completed four experimental tasks: two assessing attention (flanker and spatial Stroop), one for recognition memory, one for working memory (n-back). Following group-level difference analyses, a discriminant analysis was performed to identify which tasks, and performance metrics possessed the highest sensitivity for distinguishing PD patients from CU individuals. Results. At the group level, PD patients exhibited significantly worse conflict costs in both attention tasks and lower sensitivity scores (d') in the recognition memory task compared to CU controls. The discriminant analysis revealed that time-based measures from the spatial Stroop task and the sensitivity score from the recognition memory task provided the highest discriminating power to differentiate between the two groups. Conclusion. These findings suggest that cognitive deficits in PD can be identified with high diagnostic accuracy using a targeted subset of metrics, eliminating the need for extensive and redundant neuropsychological testing batteries for attention and memory, without needing an extensive number of cognitive tasks for attention and memory.

09.
arXiv (quant-ph) 2026-06-15

Sensitivity of polaron-molecule observables to MDR/GUP-like ultraviolet deformations at low energies via quantum computing

arXiv:2606.14479v1 Announce Type: new Abstract: We show that impurity many-body observables can display enhanced sensitivity to ultraviolet deformations of generalized-uncertainty-principle and modified-dispersion-relation type at accessible energy scales. Using a deformed polaron-molecule Hamiltonian constructed to preserve the infrared sector, we quantify the impact of such deformations on spectral and Ramsey observables and implement the corresponding dynamics in a controlled quantum computing setting. We identify regimes near the polaron-molecule crossover where small ultraviolet deformations are strongly amplified, leading to experimentally resolvable changes in quasiparticle properties and spectral response. Our results establish a concrete sensitivity-based route to low-energy quantum-gravity phenomenology in a well-defined many-body platform and delimit the validity of the effective description. Furthermore, we report experimental validation on the QRed superconducting quantum processor (BSC-CNS).

10.
medRxiv (Medicine) 2026-06-15

Efficacy of Painhunting Therapy for Event-Related Depression: A Randomized Controlled Trial with Crossover Replication

Background. Depression affects an estimated 332 million people worldwide and is a leading cause of disability, with up to 80% of major depressive episodes preceded by an identifiable adverse life event [17,18]. First-line treatments target symptoms rather than the precipitating event and are resource-intensive: standard CBT averages roughly 12 sessions, and antidepressant discontinuation carries relapse rates near 35% at six months [8]. These limitations create a clear rationale for brief, structured interventions that address the cognitive and somatic sequelae of adverse life events directly. Painhunting therapy is one such intervention, in which each session targets a discrete adverse event through a structured incident-processing procedure. Methods. We conducted a two-arm, parallel-group, single-site randomised controlled trial comparing Painhunting therapy (Arm A, immediate; n=42) with a waitlist control (Arm B, delayed; n=42) in adults with PHQ-9 >= 9 and active psychological distress related to an adverse life event. After the primary endpoint at T2 (approximately two weeks post-randomisation), Arm B crossed over to active treatment, with T3 as the post-crossover endpoint at approximately four weeks. The primary outcome was PHQ-9 at T2 (between-arm contrast); secondary outcomes were ICG, GAD-7, WHO-DAS 2.0 (12-item), and the Global Impression of Change (GIC). Pre-specified analyses included intention-to-treat, per-protocol, and single-exclusion sensitivity populations. Results. Eighty-four participants were randomised (198 applications, 134 completed screening questionnaire, 119 passed psychometric screening). At T2, mean PHQ-9 was 2.32 (SD 2.59) in Arm A and 16.56 (SD 6.76) in Arm B, yielding an ITT between-arm Cohen d = 2.78 (95% CI 2.19-3.76, p < 0.001). Within-arm paired reductions during each arm's active-treatment window reproduced this magnitude (Arm A T0 to T2 change 14.71, Morris d = 2.80; Arm B T2 to T3 change 14.19, Morris d = 2.77, eligible n=26). Treatment gains were durable at the T4 follow-up (week 8). Aligning each arm to its own end-of-treatment timepoint, the off-treatment drift to week 8 was almost identical between arms: Arm A rose 0.78 points from T2 to T4 (2.19 to 2.97, n=37) and Arm B rose 1.59 points from T3 to T4 (4.74 to 6.33, n=27), the latter falling to 0.77 points once a single documented relapse case (R59) is excluded (4.81 to 5.58, n=26). This small off-treatment rebound then stabilised rather than continuing: Arm A was essentially unchanged from T3 to T4 (change +0.05), with concordant maintenance on ICG, GAD-7, and WHO-DAS. At T4, 68% of Arm A and 41% of Arm B remained in remission (PHQ-9 < 5). Secondary measures (ICG, GAD-7, WHO-DAS) moved in the same direction and to comparable magnitude at every timepoint. The waitlist window in Arm B showed essentially no change on any measure (PHQ-9 change 0.22, p = 0.81). Sensitivity analyses excluding six sub-threshold T2 cases, the single treated-in-error case (R82), the R59 relapse case, and one late T2 submitter left all conclusions unchanged. Conclusions. Painhunting therapy produced large and statistically robust reductions in depression, complicated grief, anxiety, and functional disability over a brief course of three to four sessions, with effect sizes substantially exceeding benchmarks reported for established first-line psychotherapies including CBT and EMDR. Critically, these gains persisted at the week-8 follow-up: depression scores in the immediate-treatment arm were essentially unchanged from four weeks to eight weeks post-randomisation, indicating that the benefit reflects durable change rather than a transient post-session dip. Treatment-window concordance between arms, durability of gains at one month off-treatment, and the flat waitlist trajectory together strengthen the evidence for genuine efficacy rather than spontaneous remission. Baseline covariates including therapeutic alliance, treatment expectancy, self-efficacy, age, and sex showed near-zero associations with outcome, reducing the plausibility of allegiance bias or expectancy effects as primary drivers. The differential retention between arms (88% vs 64% at T3) is attributable to the waitlist design and is discussed as a limitation. These findings support proceeding to a confirmatory active-comparator trial against manualized CBT. Trial registration: ClinicalTrials.gov NCT07490691, prospectively registered.

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

Massive Open-Vocabulary Keyword Spotting

Automatic speech recognition systems have been shown to under-perform when it comes to transcribing words rarely seen in the training data, namely specialized terminology. Open-vocabulary keyword spotting, combined with contextual biasing, has been shown to mitigate this issue. However, existing systems can only handle glossaries of a few hundred terms without becoming an infeasible bottleneck. We propose a system that stores features with a memory footprint up to 128 times smaller than a comparable baseline and allows users to process massive databases while remaining open-vocabulary. Without fine-tuning the speech recognition model, our system achieves a comparable entity recall as uncompressed solutions, even in languages not seen during training.

13.
arXiv (CS.CL) 2026-06-15

Knowing When to Quit: A Principled Framework for Dynamic Abstention in LLM Reasoning

LLMs utilizing chain-of-thought reasoning often waste substantial compute by producing long, incorrect responses. Abstention can mitigate this by withholding outputs unlikely to be correct. While most abstention methods decide to withhold outputs before or after generation, dynamic mid-generation abstention considers early termination of unpromising reasoning traces at each token position. Prior work has explored empirical variants of this idea, but principled guidance for the abstention rule remains lacking. We present a formal analysis of dynamic abstention for LLMs, modeling abstention as an explicit action within a regularized reinforcement learning framework. An abstention reward parameter controls the trade-off between compute and information. We show that abstaining when the value function falls below this reward strictly outperforms natural baselines under general conditions. We further derive a principled and efficient method to approximate the value function. Empirical results on mathematical reasoning and toxicity avoidance tasks support our theory and demonstrate improved selective accuracy over existing methods.

14.
arXiv (CS.LG) 2026-06-11

Minimal surfaces, Knots, and Neural Networks

arXiv:2605.26234v2 Announce Type: replace-cross Abstract: A recent conjecture by Joel Fine posits a relationship between the coefficients of the HOMFLY polynomial of a knot $K$ in the 3-sphere $S^3$, and the signed count of minimal surfaces in hyperbolic 4-space $\mathrm{H}^4$ meeting the sphere at infinity at $K$, with prescribed genus and self-intersection number. In this paper, we develop a novel machine learning framework based on Physics-Informed Neural Networks (PINNs) to solve the minimal surface equation in hyperbolic space. We utilise this framework to test Fine's Conjecture by constructing near-minimal surfaces bounding various families of knots in $S^3$. Furthermore, we develop an algorithmic method to find self-intersections and compute their sign. For every knot analysed, the computationally discovered minimal surfaces and their self-intersection numbers perfectly align with the predictions of Fine's Conjecture, providing empirical evidence for it.

15.
arXiv (CS.AI) 2026-06-25

Compositional Behavioral Semantics for State Abstraction in Reinforcement Learning

arXiv:2606.25357v1 Announce Type: cross Abstract: State abstraction plays a key role in scaling reinforcement learning to complex but structured systems. In studying such systems, a wide range of behavioral structures have been studied in reinforcement learning, including value functions, invariants, bisimulation relations, and behavioral metrics. However, a general principle for determining what structures are provably preserved under state abstraction is still lacking. In this paper, we present a unified framework for defining and analyzing behavioral structures in reinforcement learning. Our framework provides a compositional way to specify behavioral semantics based on local, one-step descriptions of system dynamics. Using this framework, we establish results showing how behavioral structures can be safely transferred between abstract and concrete systems. We further show how to construct quantitative metrics from logical behavioral semantics with soundness guarantees. Together, these results provide a principled foundation for reasoning about behaviors under state abstraction in reinforcement learning and offer reusable definition and proof principles for a broad class of behavioral structures in reinforcement learning.

16.
arXiv (CS.CV) 2026-06-24

SER: Learning to Ground Video Reasoning with Semantic Evidence Rewards

Video MLLMs often struggle with fine-grained spatio-temporal reasoning, sometimes generating correct answers based on irrelevant frames or objects. Although outputting spatio-temporal evidence during reasoning is a promising direction, existing RL frameworks typically rely on geometry-only (IoU) rewards, which can be sensitive to boundary perturbations and overlook semantic alignment. To address this, we propose Semantic Evidence Reward (SER), which reformulates spatio-temporal evidence grounding as a constrained verification task. Instead of computing pixel-level overlap, SER uses a referee VLM as a local checker to evaluate model-generated evidence claims across two dimensions: relevance and localization quality, combined with a temporal penalty. This design reduces the reliance on dense box annotations and enables training directly on standard video QA data. On the V-STAR benchmark, SER achieves 49.6% mLGM, improving by 3.0 points over the strong evidence-grounded baseline Open-o3-Video, demonstrating its potential in enhancing both answer accuracy and evidence grounding.

17.
arXiv (CS.CV) 2026-06-16

Variational Deep Unfolding with Mamba-Based Nonlocal Modeling for Underwater Image Enhancement

Underwater imaging plays a crucial role in ocean engineering, although captured data often suffer from poor visibility and color distortion. To address these challenges, we propose a model-based deep unfolding network for underwater image enhancement that integrates variational modeling into a learnable architecture. The framework is guided by a variational formulation based on a dehazing decomposition, incorporating a multiplicative residual component to absorb remaining artifacts and a nonlocal gradient-type constraint to preserve structural details and enhance edge sharpness. We provide a theoretical analysis establishing the existence of solution for the associated minimization problem. The proposed unfolding method incorporates Mamba layers to efficiently capture self-similarities in the scene. In addition, we introduce a proximal trajectory loss that enforces consistency between the unfolding stages and the iterations of an ideal restoration regularizer. Experimental results demonstrate that the proposed unfolding approach achieves improved visual quality and competitive quantitative performance compared with recent state-of-the-art methods. The source code will be available at https://github.com/MIA-UIB/Variational-Unfolding-Mamba-Underwater-Enhancement .

18.
arXiv (CS.AI) 2026-06-12

Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents

arXiv:2606.10616v2 Announce Type: replace Abstract: Long-horizon language agents accumulate observations, reasoning traces, and retrieved facts that exceed their finite context windows, making memory retention a fundamental resource-allocation problem. Existing memory systems improve management through heuristic scoring, retrieval optimization, or learned compression, but largely treat retention as a local decision problem and do not explicitly model its long-term consequences under realistic observability constraints. To fill this gap, we formulate memory retention as a constrained stochastic optimization problem with explicit budget feasibility, evidence utility, and delayed costs including miss penalties, reacquisition delays, and stale-information risk. We then propose OSL-MR (Observability-Safe Learning for Memory Retention), a novel framework that enforces a strict separation between online-observable features and offline-available supervision (OAS). OSL-MR combines an evidence learner trained from realized evidence supervision with a Mixed-Score heuristic that serves both as a deployable online-safe baseline and as a structured inductive prior for learning. The resulting policy learns query-conditioned evidence value directly from interaction data while remaining deployable under the same observability constraints. Experiments on LOCOMO and LongMemEval show that OSL-MR consistently outperforms recency-based methods, Generative Agents-style scoring, and other heuristic baselines, particularly under tight memory budgets. The Mixed-Score prior further improves precision while preserving recall, and sensitivity analysis demonstrates robustness across a wide range of cost configurations.

19.
arXiv (CS.LG) 2026-06-12

Multimodal Graph Negative Learning

arXiv:2606.12863v1 Announce Type: new Abstract: Multimodal attributed graphs (MAGs) integrate graph topology with heterogeneous modality attributes, such as text and images, thereby enabling richer modeling of complex relational systems. However, such expressiveness also makes learning on MAGs depend on multiple semantic sources, including structural topology, textual and visual attributes, each of which can be regarded as a branch for node representation. Node-level branch semantic imbalance arises when these branches differ across nodes in semantic informativeness and reliability: a branch that provides discriminative semantics for one node may mislead another due to bias in modality quality or structural context. Existing methods often mitigate such heterogeneity through cross-branch agreement or alignment, implicitly treating the dominant prediction as reliable supervision. When the dominant branch is biased, forced imitation may propagate its bias to other branches and suppress original semantics that are useful for classification. We propose GraphMNL, a graph-aware multimodal negative learning framework that addresses this issue by using Negative Learning as cross-branch guidance. Instead of forcing inferior branches to imitate a teacher prediction, the model teaches them which classes a node is unlikely to belong to. GraphMNL builds a branch library, identifies dominant and inferior branches via graph-aware reliability arbitration, gates unstable transfer, and applies target-preserving negative learning over non-target classes. This design decouples target supervision from branch guidance so that supervised losses learn the correct class, while Negative Learning suppresses unlikely alternatives when branch agreement is unreliable. Through the comprehensive experimental evaluation, GraphMNL achieves the best performance on Grocery datasets with 72.47% accuracy and 76.60 F1 score on Reddit M datasets.

20.
arXiv (CS.AI) 2026-06-19

KG-SoftMAP: Soft Knowledge-Graph Priors for Bayesian Network Structure Learning from Sparse Discrete Data

arXiv:2606.10358v2 Announce Type: replace-cross Abstract: Learning Bayesian network (BN) structure from sparse discrete data is hard: when each instance records only a few variables, most variable pairs lack the joint observations needed for reliable scoring, and data-only methods recover little structure. However, imperfect domain knowledge, expressible as a weighted directed knowledge graph (KG), is often available. We propose KG-SoftMAP, which encodes such a KG as a finite-strength, confidence-weighted edge prior and maximizes a MAP objective combining the BDeu score with a logit-form prior; the KG may be expert-curated or LLM-extracted. On synthetic benchmarks with known DAGs, KG-SoftMAP reaches Directed-F1 (DF1) $0.19$–$0.32$ at observation rate $\rho=0.05$ and DF1 $0.44$–$0.97$ at $\rho\geq0.2$, while every data-only learner tested stays near zero under the same sparse masks. Recovery tracks KG quality: controlled corruption degrades it smoothly, a zero-signal KG yields DF1 $0.00$, and a blindly LLM-extracted KG with imperfect precision and recall still drives substantial recovery. On three real sparse educational datasets, the learned BN acts as a concept-level posterior model: on SAF it matches logistic regression (LR) within $0.03$ F1_FAIL while providing an inspectable concept graph, calibrated Fail probabilities, and tractable posterior queries from partial observations.

21.
arXiv (CS.CL) 2026-06-17

PromptMN: Pseudo Prompting Language

Prompting has become the primary interface between humans and generative AI, yet many natural language prompts remain fragile: roles, goals, constraints, and expected outputs are often buried in prose or left implicit. In agentic and software development workflows, a misread at the first handoff can propagate through every step, since a significant portion of agent failures stem from context ambiguities rather than model limitations. This paper introduces PromptMN, a pseudo-prompting domain-specific language that annotates natural language with compact, %-prefixed typed directives covering roles, goals, requirements, priorities, constraints, plans, inputs, and outputs. Semantic resolution lets authors write in any order while the model interprets directives by function. PromptMN sits between informal prompting and programming-style pseudocode: structured enough to be inspectable and reusable, yet lightweight enough for analysts, managers, developers, and stakeholders across the software development lifecycle (SDLC). PromptMN also pairs with reverse prompt engineering. Asking a model to restate a desired outcome as PromptMN lets users inspect the inferred roles, goals, constraints, and missing assumptions before acting, reducing repair cycles and yielding a reusable artifact for aligning people and AI tools. PromptMN's feasibility is evaluated across several frontier models, including Claude Fable 5, Claude Opus 4.8, Gemini 3.1 Pro, and GPT-5.5. The models correctly resolved PromptMN instructions, including complex structures such as repetition, conditionals, methods, and a prime-checking task, without fine-tuning. The same vocabulary applies across new codebases, maintenance, and redesign in the SDLC scenarios presented. While large-scale validation remains future work, these early results suggest PromptMN is a practical step toward clearer, more reviewable human-to-AI interaction.

22.
arXiv (CS.CL) 2026-06-12

InnoEval: On Research Idea Evaluation as a Knowledge-Grounded, Multi-Perspective Reasoning Problem

The rapid evolution of Large Language Models has catalyzed a surge in scientific idea production, yet this leap has not been accompanied by a matching advance in idea evaluation. The fundamental nature of scientific evaluation needs knowledgeable grounding, collective deliberation, and multi-criteria decision-making. However, existing idea evaluation methods often suffer from narrow knowledge horizons, flattened evaluation dimensions, and the inherent bias in LLM-as-a-Judge. To address these, we regard idea evaluation as a knowledge-grounded, multi-perspective reasoning problem and introduce InnoEval, a deep innovation evaluation framework designed to emulate human-level idea assessment. We apply a heterogeneous deep knowledge search engine that retrieves and grounds dynamic evidence from diverse online sources. We further achieve review consensus with an innovation review board containing reviewers with distinct academic backgrounds, enabling a multi-dimensional decoupled evaluation across multiple metrics. We construct comprehensive datasets derived from authoritative peer-reviewed submissions to benchmark InnoEval. Experiments demonstrate that InnoEval can consistently outperform baselines in point-wise, pair-wise, and group-wise evaluation tasks, exhibiting judgment patterns and consensus highly aligned with human experts.

23.
arXiv (CS.CL) 2026-06-15

Be My Tutor: On-Policy Co-Distillation for Mutual LLM Improvement via Peer Feedback

We study multi-domain LLM training in which two models, each stronger in a different domain, co-evolve by tutoring each other through on-policy feedback. Unlike one-way distillation or single-model fine-tuning, our goal is mutual Pareto improvement: each model improves across domains without losing its original strength. To this end, we propose On-Policy Co-Distillation (OPCoD), where each student's self-distillation is conditioned on its own correct rollout and feedback from its peer. To make feedback exchange effective, OPCoD uses cognizance-based gating to decide when to give feedback and feedback anchoring to ground feedback in the problem. On Science Q\&A tasks, OPCoD consistently outperforms baselines and achieves Pareto improvement across all evaluated domain pairs and students.

24.
arXiv (CS.CL) 2026-06-24

Agon: An Autonomous Large-Scale Omnidisciplinary Research System Built on Prompt Economy

Large language models are making research production scalable, shifting the bottleneck from producing artifacts to judging claims. We present \textsc{Agon}, a research orchestrator that validates what can be checked inside the workflow and leaves the remaining judgments to human scientists. \textsc{Agon} is built on six design principles: Prompt Economy, Future-Facing, Minimal Prompts, OmniDisciplinary, Massive Parallelism, and Zero-Code. We ran \textsc{Agon} across domains for 444 iterations of Prompt Economy loops, using only small starting topics and no human-written experimental code. These deployments demonstrate scalability while exposing new classes of failure. We organize these failures into a taxonomy along severity, fixability, visibility, and capability locus. The taxonomy separates failures the loops can see and fix from those that require human judgment. Together, these results show that \textsc{Agon} is pushing research toward a new paradigm: machine scales, human steers.

25.
arXiv (CS.LG) 2026-06-18

Decomposing Prediction Mechanisms for In-Context Recall

arXiv:2507.01414v2 Announce Type: replace Abstract: We introduce a new family of toy problems that combine features of linear-regression-style continuous in-context learning (ICL) with discrete associative recall. We pretrain transformer models on sample traces from this toy, specifically symbolically-labeled interleaved state observations from randomly drawn linear deterministic dynamical systems. We study if the transformer models can recall the state of a sequence previously seen in its context when prompted to do so with the corresponding in-context label. Taking a closer look at this task, it becomes clear that the model must perform two functions: (1) identify which system's state should be recalled and apply that system to its last seen state, and (2) continuing to apply the correct system to predict the subsequent states. Training dynamics reveal that the first capability emerges well into a model's training. Surprisingly, the second capability, of continuing the prediction of a resumed sequence, develops much earlier. Via out-of-distribution experiments, and a mechanistic analysis on model weights via edge pruning, we find that next-token prediction for this toy problem involves at least two separate mechanisms. One mechanism uses the discrete symbolic labels to do the associative recall required to predict the start of a resumption of a previously seen sequence. The second mechanism, which is largely agnostic to the discrete symbolic labels, performs a "Bayesian-style" prediction based on the previous token and the context. These two mechanisms have different learning dynamics. To confirm that this multi-mechanism (manifesting as separate phase transitions) phenomenon is not just an artifact of our toy setting, we used OLMo training checkpoints on an ICL translation task to see a similar phenomenon: a decisive gap in the emergence of first-task-token performance vs second-task-token performance.