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

Weibull Weight-Scale Parameter Evolution under AdamW Training Dynamics

作者:

arXiv:2606.19367v1 Announce Type: new Abstract: Building on a two-parameter Weibull framework for diagnosing transformer weight distributions, we study why the Weibull weight-scale parameter $\lambda$ grows, overshoots, and then relaxes during AdamW training. We derive a leading-order three-force decomposition of the squared weight norm from the AdamW update: an alignment force measuring the correlation between weights and the adaptive update direction, an injection force from adaptive step magnitude, and a decay force from decoupled weight decay. On self-trained Pythia-70M models with ground-truth optimizer moments, alignment dominates the rise phase, contributing 88-94% of the absolute force budget across four random seeds and remaining robust to super-weight removal. Near saturation, alignment and decay approach balance, explaining the transition from weight-scale growth to relaxation. These force dynamics directly govern the squared-norm component underlying $\lambda(t)$; the remaining RMS-to-Weibull reconstruction offset is measurable and decomposes into bridge and integration components, totaling approximately 5-6% in densely sampled regions. To extend the analysis to real models where optimizer moments are unavailable, we introduce a spline displacement method that recovers the alignment force from sparse checkpoints with approximately 92-94% accuracy, about twice the naive two-point baseline. We further observe that the peak value of $\lambda(t)$ varies with training-data coherence in our experiments, suggesting a data-dependent component of weight-scale growth that we leave to a controlled follow-up study. Code and data are available at https://github.com/tiexinding/NPM-Weibull-public.

02.
arXiv (CS.AI) 2026-06-16

Inference-time Policy Steering via Vision and Touch

arXiv:2606.14981v1 Announce Type: cross Abstract: Inference-time steering adapts pre-trained generative robot policies during deployment by verifying candidate actions before execution. While prior methods typically perform this verification only with visual observations, vision alone is often insufficient for contact-rich manipulation, where success depends on both global task progress and subtle local interactions such as contact force. We introduce ViTaL, a visuo-tactile inference-time steering framework that formulates multimodal guidance as a bi-level optimization problem. At the high level, visual sampling-and-verification performs long-horizon mode selection, deciding what behavior the robot should execute. At the low level, tactile-guided diffusion editing refines the selected action sequence over a shorter horizon to satisfy local contact requirements. To support outcome-based steering, ViTaL learns a visuo-tactile latent world model and employs semantically aligned visual and tactile verifiers, including a novel text-conditioned tactile reward that scores predicted tactile futures directly in latent space. Across three real-world contact-rich manipulation tasks, ViTaL improves overall success by 51% over the base policy, outperforms unimodal steering by at least 33%, and exceeds naive multimodal fusion by at least 20%. Website: https://yilin-wu98.github.io/vital_website.

03.
arXiv (quant-ph) 2026-06-24

A Universal All-Fiber Quantum Buffer for the Telecom Band

arXiv:2606.24681v1 Announce Type: new Abstract: The realization of a scalable quantum internet relies on the ability to temporally align asynchronous photonic signals through on-demand buffering. While matter-based quantum memories achieve long storage times, their extremely narrow bandwidths and cryogenic requirements pose significant barriers to integration with existing telecommunications infrastructure. Conversely, current all-optical memories operate at room temperature but are hampered by high input/output losses and a lack of universality across different photonic degrees of freedom. Here, we demonstrate a universal, fully fiber-integrated quantum buffer operating over the full telecom C-band that overcomes these fundamental trade-offs. By implementing an actively switched dual-Sagnac cavity driven by cross-phase modulation, we achieve an ultra-low input/output loss of 0.46 dB and a storage time exceeding 18 $\mu$s. The device exhibits an operational bandwidth exceeding 12.5 THz ($\sim$100 nm), covering the full telecom C-band. We show the simultaneous buffering of over 200 temporal modes with the ability to address them either collectively or one by one. We demonstrate high-fidelity storage for all three degrees of freedom compatible with optical fiber propagation, namely time-bin, frequency-bin, and polarization qubits, along with faithful preservation of entanglement, confirming the platform's true universality. These results provide a robust, room-temperature solution for the high-rate synchronization of multidimensional quantum states, clearing a major hurdle for the deployment of global photonic quantum networks.

04.
arXiv (CS.AI) 2026-06-16

An AI Security Agent for University ACMIS: Multi-Vector Threat Detection and Automated Response

arXiv:2606.08270v2 Announce Type: replace-cross Abstract: University Academic Management Information Systems (ACMIS) are high-value targets for a wide spectrum of security threats including brute-force login attacks, payment fraud, privilege escalation, insider data theft, and academic integrity violations. Traditional rule-based intrusion detection systems are inadequate because many malicious activities are structurally indistinguishable from normal operations. This paper presents an AI-based security agent for ACMIS that combines supervised anomaly detection, behavioural analytics, and a natural language processing chatbot for secure password recovery. The agent monitors five operational layers: authentication, authorisation, financial transactions, user behaviour, and system health, and responds through a four-tier risk escalation framework. A modular architecture allows the core engine to be extended to other institutional systems. Experiments on a simulated ACMIS event log dataset of 147,922 sessions demonstrate a threat detection macro-average F1 of 0.966, compared to 0.156 for a rule-based baseline and 0.836 for a sequence-only (LSTM) baseline, with end-to-end critical-tier automated response latency under 1 ms on a single-node prototype. The integrated recovery chatbot achieves 97.1 percent identity verification accuracy and an 87.3 percent mass-reset attack detection rate with zero false positives on legitimate high volume recovery periods.

05.
arXiv (CS.AI) 2026-06-17

Shattering the Autoregressive Curse: Dynamic Epistemic Entropy Orchestrated Erasable Reinforcement Learning for LLMs

arXiv:2606.17735v1 Announce Type: new Abstract: Although reinforcement learning (RL) has expanded the cognitive boundaries of large language models (LLMs), it often remains vulnerable to the autoregressive curse in long-horizon logical reasoning: small epistemic perturbations introduced early in generation can propagate irreversibly along the Markov decision process flow, triggering cascading failures that drive the reasoning trajectory toward collapse. To overcome this autoregressive cascade, in which a single early mistake can compromise all subsequent reasoning steps, we propose dynamic epistemic entropy orchestrated erasable reinforcement learning ($E^3RL$). $E^3RL$ eliminates reliance on external signals by grounding the model's endogenous local autoregressive cross-entropy as an intrinsic coordinate of epistemic uncertainty. By introducing segment-level adaptive dynamic thresholds and advantage allocation, $E^3RL$ enables the model to precisely excise localized logical defects while reusing historical key-value (KV) cache streams, thereby endowing the reasoning process with a self-healing capability. We train $E^3RL$ on the DeepMath-103k dataset. Experimental results show that $E^3RL$ reshapes the exploration efficiency of long-sequence reasoning and improves sample efficiency while maintaining linear memory overhead. On mathematical reasoning benchmarks such as AIME, $E^3RL$ achieves substantial performance gains, with the 4B and 8B parameter models surpassing previous state-of-the-art (SOTA) results by 5.349\% and 6.514\%, respectively. These findings suggest that $E^3RL$ shatters the autoregressive curse in long-sequence reasoning and establishes a theoretical and systems-level foundation for the next generation of self-healing artificial general intelligence (AGI).

06.
bioRxiv (Bioinfo) 2026-06-12

Deciphering cross-omics complexity of tissues via diagonal integration of unpaired spatial multi-omics data

Recent spatial multi-omics technologies enable the simultaneous in situ profiling of multiple omics modalities on the same tissue section; however, they face challenges in experimental complexity and high costs. This technical limitation can be circumvented by diagonal integration methods, which integrate omics data from different modalities. However, existing single-cell diagonal integration approaches overlook spatial information, causing unreliable anchoring across omics layers. Here, we introduce STAMO, a graph attention neural network model for spatially aware integration of unpaired spatial slices from different omics. Systematic benchmarking on spatial epigenome-transcriptome slices proves that STAMO outperforms the state-of-the-art methods in generating aligned embeddings and identifying consensus spatial domains across omics. We apply STAMO to integrate unpaired data from diverse spatial omics types (transcripts, epigenetics, DNA, and proteins), including slices from spatial RNA and four different epigenomic modalities, spatial ATAC and RNA slices across embryonic stages, spatial protein and RNA slices, and spatial DNA and RNA slices. In addition, the integration capability of STAMO can be further used to achieve cross-omics generation, offering a solution for exploring spatial region-specific gene regulatory mechanisms.

07.
arXiv (CS.LG) 2026-06-24

Layer-wise Geometric Approximation Rates for Deep Networks

arXiv:2604.20219v2 Announce Type: replace Abstract: Depth is widely viewed as a central contributor to the success of deep neural networks, whereas standard neural network approximation theory typically provides guarantees only for the final output and leaves the role of intermediate layers largely unclear. We address this gap by developing a quantitative framework in which depth admits a precise scale-dependent interpretation. Specifically, we design a single shared mixed-activation architecture of fixed width $2dN+d+2$ and any prescribed finite depth such that each intermediate readout $\Phi_\ell$ is itself an approximant to the target function $f$. For $f\in L^p([0,1]^d)$ with $p\in [1,\infty)$, the approximation error of $\Phi_\ell$ is controlled by $(2d+1)$ times the $L^p$ modulus of continuity at the geometric scale $N^{-\ell}$ for all $\ell$. The estimate reduces to the geometric rate $(2d+1)N^{-\ell}$ if $f$ is $1$-Lipschitz. Our network design is inspired by multigrade deep learning, where depth serves as a progressive refinement mechanism. For every prescribed terminal depth, the construction yields a finite nested family of prefix readouts whose earlier correction terms remain embedded in later readouts. Thus the approximation may be truncated within the prescribed depth range once the desired certified accuracy is reached.

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

How Far Can Machine Translation Quality Take You? Extrinsic Discourse Evaluation in Goal-Oriented Setups

Existing machine translation (MT) metrics and discourse-focused evaluations primarily assess translation quality intrinsically, without measuring the downstream consequences of translation errors. In this work, we focus on extrinsic discourse evaluation of machine translation under two distinct regimes: static and interactive. Under the static regime, we propose an entity counting task as a probe of referential consistency in discourse. We show that high intrinsic MT quality does not reliably predict downstream discourse success and strong MT systems still produce referential inconsistencies. For the interactive regime, we study the goal-oriented multi-agent Welfare Diplomacy game as a probe of long-horizon communication and coordination. We find that interaction-specific translation failures impact downstream coordination. Our results highlight goal-oriented environments as a viable framework for discourse-sensitive extrinsic MT evaluation.

09.
arXiv (CS.AI) 2026-06-16

MR-GVNO: A Geometry-Aware Variational Physics-Informed Neural Operator for Mindlin-Reissner Plates on Irregular Domains

arXiv:2606.16624v1 Announce Type: new Abstract: Plate and shell structures are widely used in engineering, making rapid response prediction under varying geometries, materials, and loads highly desirable. However, conventional finite element methods require repeated modeling and solution, resulting in high computational costs. This study proposes a geometry-aware variational neural operator for Mindlin-Reissner plate problems, termed MR-GVNO. The method uses boundary point clouds to represent irregular geometries and employs separate encoders for spatially varying material fields, pressure loads, and scalar physical parameters. A cross-attention mechanism integrates these inputs with query point information to predict transverse deflections and rotations at arbitrary locations. MR-GVNO is trained without labeled solution data using a variational physics-informed loss derived from the discretized total potential energy. It directly processes irregular point clouds and allows different physical fields to be discretized independently, avoiding interpolation onto a common grid. Numerical experiments on single-hole, double-hole, and L-shaped plates demonstrate accurate response prediction under homogeneous and heterogeneous materials and uniform and random loads. The model also achieves millisecond-level full-field inference and favorable cross-geometry generalization.

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

ViPER: Vision-based Packing-Aware Encoder for Robust Malware Detection

Visualization-based malware detection maps raw binary bytes to grayscale images and applies learned visual classifiers, providing an evasion-resistant and disassembly-free alternative to conventional analysis pipelines. However, executable packing remains a critical failure mode: packed binaries produce high-entropy images that obscure the structural patterns these models rely on. Because packing is also prevalent in benign software (e.g., for compression or copy protection), packing state alone is not a reliable indicator of maliciousness, and existing approaches do not address this challenge within a unified supervised framework. We present ViPER, a Vision-based Packing-Aware Encoder for Robust malware detection. ViPER builds on a LoRA-adapted ViT-B/14 backbone with a dual-head architecture that jointly learns malware classification and packing detection. A packing-aware gating mechanism conditions malware predictions on the inferred packing state, enabling distinct decision boundaries for packed and unpacked inputs. To address packing label skew during training, we employ frequency-weighted losses with stratified sampling over joint class-packing strata. Evaluated on 200,000 Windows PE byteplot images, ViPER achieves a balanced accuracy of 0.8521, ROC-AUC of 0.9260, and AUPR of 0.9279, outperforming representative state-of-the-art baselines across all primary metrics, while attaining a packing detection AUC of 0.9949.

11.
arXiv (CS.AI) 2026-06-16

Auditing Reward Hackability in Code RL Training Environments

arXiv:2606.16062v1 Announce Type: new Abstract: We measure the rate at which code RL environments accept incorrect solutions as correct. On a 49-task sample of SWE-bench Verified, 28.5% of tasks have test suites weak enough that a Docker-verified incorrect patch passes them. On 20 R2E-Gym tasks across 6 repositories, the same pipeline at single-shot exploit generation yields 25.0%. A random-effects meta-analysis over 134 frontier model submissions to SWE-bench Verified finds, within the same human-rated difficulty stratum, model Pass@1 is +14.14 percentage points higher on flagged-hackable tasks than on robust ones (95% CI [+11.80, +16.48]; one-sided p < 10^-6; I^2 = 0%; 123 of 134 models positive). We then describe a procedure for hardening the broken tasks. An inline LLM judge with a Docker gold-sanity gate runs each generated test against the gold solution before the judge is consulted. On the 11 broken tasks in the audit, the gate flags 65 of 105 decisive LLM-generated tests as failing on the gold patch itself, a 61.9% per-augmentation defect rate the LLM judge alone misses. With diversity-biased retry, the loop converges 9 of 11 tasks to a gated upgrade.

12.
arXiv (CS.LG) 2026-06-16

ExpRL: Exploratory RL for LLM Mid-Training

arXiv:2606.17024v1 Announce Type: new Abstract: Sparse reward reinforcement learning (RL) has become a standard tool for improving LLM reasoning, but its success depends critically on the coverage present in the base model. In practice, models are often primed for RL through mid-training on curated reasoning traces that teach useful primitive skills such as decomposition, verification, or self-correction. Although effective, this strategy requires manually specifying what the model should learn, and it remains unclear whether such primitive coverage is enough for much harder problems, which require combining these skills into broader solution strategies. We study a more automated approach: RL-based mid-training using large corpora of human-written question-answer data. Rather than treating reference solutions as targets to imitate, our method, ExpRL, uses them as reward scaffolds: references are hidden from the policy and used only to construct problem-specific grading rubrics for judging on-policy reasoning traces. The policy samples from the original problem prompt, while an LLM judge compares the sampled reasoning trace against the reference solution and assigns outcome-level or process-level dense rewards. This lets ExpRL reinforce partial progress, useful intermediate reductions, and productive reasoning behaviors that sparse final-answer rewards often fail to upweight. On challenging math reasoning tasks, ExpRL yields stronger RL priming than SFT, sparse-reward GRPO, and self-distillation, and provides a better initialization for subsequent sparse-reward RL. Additional mixed-domain experiments further suggest that ExpRL can extend beyond the original math-only setting.

13.
arXiv (quant-ph) 2026-06-19

Indefinite Quantum Causality

arXiv:2606.19438v1 Announce Type: new Abstract: In recent years, operational approaches to quantum foundations have been developed as a means of understanding the core principles and distinctive features of quantum theory. Such approaches typically view physical processes as sequences of operations, with earlier operations serving as causes of later effects. However, a growing literature is emerging on the possibility of relaxing this assumption and allowing for quantum indefiniteness in the causal order. This development stems from a variety of motivations, both fundamental and applied, including exploring the role of causality in quantum theory, the interplay between quantum theory and general relativity, and higher-order quantum computing. A prominent offshoot of this development is the emergence of indefinite causal order as a feasible resource for quantum information processing. This review provides an overview of the current state of the art in the field, covering the methodology underlying indefinite quantum causality within the so-called "process matrix formalism", outlining key results and experimental implementations, and discussing recent advances.

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

Battery detection of XRay images using transfer learning

The need for detecting and sorting batteries is drastically increasing for many applications. This study proves the potential of transfer learning in predicting whether the image contains a battery or not, the location and identifying three types of batteries, namely: prismatic, pouch, and cylindrical Lithium-Ion Batteries (LIB). Particularly, it focuses on the transfer learning method in two applications: Training a large-scale dataset to detect electronic devices using a pre-trained YOLOv5m, then using these latter trained weights to detect and classify the batteries. The precision of battery detection achieves 94%, which outperforms the pretrained YOLOv5m weights with 5%, in 22 ms inference time.

15.
arXiv (CS.AI) 2026-06-17

Learn to Quantify Social Interaction with Constraints for Pedestrian Walking

作者:

arXiv:2606.17897v1 Announce Type: new Abstract: Long-term human path forecasting in crowds is critical for autonomous moving platforms (like autonomous driving cars and social robots) to avoid collision and make high-quality planning. Although the current research take into account social interactions for prediction, they don't reveal the exact kinds of social interactions happened among people and how the social interactions affect the decision-making process of pedestrians, which further limits its robustness. Social interactions in pedestrian walking are intuitively massive and hard to label and quantify. In this paper, we explore creatively to quantify and interpret how pedestrians interact with others by proposing Learn to Cluster. Our clustering social interactions is probabilistic latent variable generative, learning directly from sequential trajectory observations, scalable to arbitrary number of pedestrians. Learn to cluster is label-free and can be naturally integrated into the training process of the prediction model. The latent variables will then serve as 'labels' to categorize social interactions. Extensive experiments over several trajectory prediction benchmarks demonstrate that our method is able to learn the patterns of social interactions and effectively integrate the patterns to pedestrian trajectory prediction.

16.
arXiv (CS.CV) 2026-06-12

RepWAM: World Action Modeling with Representation Visual-Action Tokenizers

This work presents RepWAM, a representation-centric world action model (WAM) built on representation visual-action tokenizers. Existing WAMs typically inherit reconstruction-oriented video tokenizers from pretrained video generation models. Although these tokenizers preserve visual fidelity, pixel reconstruction alone provides limited guidance for learning instruction-following dynamics that connect future prediction with robot control. To address this, we explore a semantic visual-action latent space for representation-centric world action modeling. Specifically, we train a representation visual-action tokenizer that maps visual inputs into aligned visual and latent action tokens. We then pretrain our WAM to jointly model future visual states and the latent actions that connect them under language instructions, followed by adaptation to real robot trajectories for closed-loop manipulation. Experiments on real-world manipulation tasks and simulation benchmarks show that RepWAM delivers strong performance across diverse manipulation settings, while ablations highlight the value of semantic visual-action tokenization over reconstruction-oriented alternatives. These results establish representation visual-action tokenization as a promising foundation for world action models and a step toward generalist robot policies. Code and weights will be available at https://github.com/wdrink/RepWAM.

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

ToaSt: Token Channel Selection and Structured Pruning for Efficient ViT

Vision Transformers (ViTs) have achieved remarkable success across various vision tasks, yet their deployment is often hindered by prohibitive computational costs. While structured weight pruning and token compression have emerged as promising solutions, they suffer from prolonged retraining and inter-layer dependencies that complicate optimization, respectively. We propose ToaSt, a decoupled framework applying specialized strategies to distinct ViT components. We apply coupled head-wise structured pruning to Multi-Head Self-Attention modules, leveraging attention operation characteristics to enhance robustness. For Feed-Forward Networks (over 60% of FLOPs), we introduce Token Channel Selection (TCS), a training-free method that filters redundant noise channels at inference time. Extensive evaluations across nine diverse models, including DeiT, ViT-MAE, and Swin Transformer, demonstrate that ToaSt achieves superior trade-offs between accuracy and efficiency, consistently outperforming existing baselines. On ViT-MAE-Huge, ToaSt achieves 88.52% accuracy (+1.64%p) with 39.4% FLOPs reduction. ToaSt also transfers effectively to diverse downstream tasks (COCO detection, ADE20K segmentation, CIFAR-100 classification), achieving 52.2 versus 51.9 mAP on COCO. Code: github.com/SHANNonLab-HUFS/ToaSt

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

Helping Figures Tell their Story! Paper-Grounded Video Generation Explaining Complex Scientific Figures

Scientific figures compress complex pipelines into a single canvas, yet understanding them requires paper-grounded, step-by-step narration aligned with visual highlights a capability missing from current video generation systems and benchmarks. To address this, we introduce paper-grounded figure-to-video generation: generating narrated, region-grounded walkthrough videos from a figure and its paper. We propose MINARD (Multimodal Interpretation of Narrated Architecture via Region Decomposition), a pipeline that generates paper-grounded narrations and sequentially grounds them to figure regions. We also release FigTalk, a benchmark with new sequential and component-level grounding metrics derived. On FigTalk, MINARD generates humanlike, paper-faithful narrations and outperforms narration-conditioned figure spatial grounding compared to existing approaches in both automatic and human evaluation

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

SCR-Guided Difficulty-Aware Optimization for Infrared Small Target Detection

Infrared small target detection remains challenging due to severe background clutter, low contrast, and weak spatial responses where geometric overlap alone is insufficient to characterize detection quality. In this work, we propose REEM (Reweighted Explicit-visibility Enhanced Modulation), a lightweight SCR-guided difficulty-aware optimization framework that incorporates Signal-to-Clutter Ratio (SCR) as a physically meaningful visibility prior during training. Instead of modifying the network architecture or directly optimizing SCR, REEM computes a ground-truth local SCR from the input image and applies a differentiable modulation to the soft-IoU learning signal, emphasizing low-visibility targets while preserving stable optimization and identical inference behavior. REEM is integrated into a U-Net-based MSHNet without introducing additional parameters, architectural modifications, or inference-time overhead. Extensive experiments demonstrate consistent improvements over the baseline, achieving higher IoU and detection probability (Pd) together with substantially reduced false alarms (FA), particularly under challenging low-visibility conditions. These results suggest that SCR-guided difficulty-aware optimization provides an effective and physically grounded complement to conventional overlap-based objectives for infrared small target detection. The code is available at https://github. com/yall-in-one/Reemm.

20.
arXiv (CS.CL) 2026-06-24

MMed-Bench-IR: A Heterogeneous Benchmark for Multilingual Medical Information Retrieval

Retrieval-augmented generation (RAG) in clinical settings increasingly requires multilingual retrieval against predominantly English evidence corpora. Multilingual medical retrieval demands three capabilities: cross-lingual alignment, concept discrimination, and evidence retrieval. However, existing benchmarks evaluate these only in isolation, leaving the interaction between biomedical expertise and multilingual coverage unmeasured. We introduce MMed-Bench-IR, a benchmark designed to disentangle these axes across 6 languages and three structurally heterogeneous tasks: (1) cross-lingual medical QA retrieval with 6,127 queries grounded in the Unified Medical Language System (UMLS), (2) concept discrimination over 4,975 confusion sets at three difficulty tiers, and (3) multilingual evidence retrieval for RAG with 2,040 quality-assured queries. The three tasks share zero concept and query overlap by design, ensuring that aggregate scores reflect genuine capability breadth. Evaluation of ten systems across six paradigm families reveals severe cross-lingual failure: biomedical encoders that score 0.818 nDCG@10 in English drop to 0.056 in Japanese, a gap that English-only benchmarks cannot detect.

21.
arXiv (CS.CL) 2026-06-18

Learning User Simulators with Turing Rewards

Learning to simulate human users in interactive settings could advance the training of agent assistants, evaluation of personalization systems, research in the social sciences, and more. Existing approaches generally do so by training a large language model (LLM) to match a single ground truth response, either by maximizing the log probability or by using a similarity reward. We instead propose {Turing-RL}: a Turing-Test-based reinforcement learning approach for training user simulator models. {Turing-RL} uses a discriminative Turing reward with an LLM judge to score how indistinguishable a generated response is from the real user's given the user's history, and the user simulator LLM learns to produce responses indistinguishable from what the user could have said with such rewards. Across two different domains–conversational chat and Reddit forum discussion–we find that {Turing-RL} consistently outperforms baseline methods on both LLM and human evaluation metrics. Our study suggests that optimizing for indistinguishability, rather than response matching, is effective for learning user simulators.

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

QMFOL: Benchmarking Large Language Model Reasoning via Quantifiable Monadic First-Order Logic Test Case Generation

arXiv:2606.20227v1 Announce Type: new Abstract: Large Language Models (LLMs) have made significant progress in reasoning, particularly in deductive reasoning, which is crucial for high-stakes decision-making. As models improve, evaluation benchmarks should evolve to keep pace. However, existing benchmarks lack fine-grained control over logical complexity and struggle to balance semantic diversity with logical consistency. To address these issues, we propose QMFOL, an automated framework for generating monadic first-order logic reasoning tasks with quantifiable and controllable complexity. It constructs formal logical structures using conjunction and disjunction patterns, enabling precise control over reasoning depth, width, label types, and distractors. These structures are then translated into natural language via LLMs, with logical consistency ensured through round-trip verification using an external prover. Based on our framework, we build QMFOLBench, a benchmark comprising 2880 instances with 960 configurations across diverse logical and semantic dimensions. Evaluations on six large reasoning models (LRMs) and two LLMs show that performance degrades and computational overhead increases with rising logical complexity. Models perform better on True-labeled tasks than on False or Unknown ones, and exhibit sensitivity to semantic variation. Overall, QMFOL offers a scalable and reliable approach for constructing deductive reasoning benchmarks with controllable complexity, enabling more precise evaluation of reasoning capabilities in modern language models.

24.
arXiv (CS.LG) 2026-06-11

Trajectory Geometry of Transformer Representations Across Layers

arXiv:2606.09287v2 Announce Type: replace Abstract: Understanding how transformer representations evolve across layers, not merely what they encode, remains an open problem in mechanistic interpretability. We recast the transformer forward pass as a discrete population trajectory through a high-dimensional representation manifold, drawing on geometric tools from computational neuroscience. Rather than probing for pre-specified features, we characterize trajectory geometry using five metrics computed directly in the ambient space: trajectory length, curvature, a semantic convergence index, layerwise cosine similarity, and representational stability. Across three model families (GPT-2, TinyLlama, Qwen2.5) and five controlled prompt families, we report four findings. First, semantically related prompts converge significantly in middle-to-late layers (peak CI 0.41–0.58, p

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

Integrating Reasoning and Generalization in Text-to-SQL via Self-Enhanced Fine-Tuning

arXiv:2606.15598v1 Announce Type: new Abstract: Text-to-SQL aims to translate natural language questions into executable SQL queries over structured databases, enabling non-expert users to access data intuitively. While recent advances in large language models (LLMs) have shown promise in this task, existing LLM-based approaches often struggle to strike a balance between strong reasoning capabilities and robust generalization. To address these limitations, we propose CoTE-SQL to enhance the LLM-based text-to-SQL generation with three key innovations: (i) self-enhanced reasoning traces distilled from LLMs without human annotation, (ii) structured chain-of-thought (CoT) prompting with modular decomposition and examples retrieval, and (iii) error-aware revision based on SQL execution feedback. Extensive experiments on the Spider and Bird benchmarks demonstrate that CoTE-SQL achieves new state-of-the-art performance among methods built on open-source LLMs with comparable model sizes on Bird (53.39% EX / 59.02 VES) and strong results on Spider (79.60% EX / 77.19 VES), with especially significant gains on complex queries. Results highlight the effectiveness of combining self-enhancement, structured reasoning, and execution-time feedback within an LLM-based framework for text-to-SQL design.