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.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.

02.
arXiv (CS.AI) 2026-06-19

RIVET: Robust Idempotent Voice Attribute Editing

arXiv:2606.19629v1 Announce Type: cross Abstract: Voice attribute editing models modify characteristics such as age and gender while preserving speaker identity. In large-scale speech datasets, however, attribute annotations are often noisy or inconsistent, which can cause conditional generative models to produce unstable edits. In this work, we show that idempotency provides an effective mechanism for improving robustness to noisy labels. An idempotent operator is one for which repeated application does not change the result, i.e., f(f(x)) = f(x). Enforcing this property acts as an implicit regularizer that reduces sensitivity to mislabeled examples. We introduce RIVET, a training framework that incorporates an idempotency objective to improve robustness to label noise. We evaluate RIVET under controlled label noise and on the GLOBE dataset with naturally noisy annotations. RIVET improves editing success and better preserves speaker identity than standard training, showing that idempotency improves robustness in voice editing models.

03.
arXiv (CS.CL) 2026-06-16

Beyond Monolingual Deep Research: Evaluating Agents and Retrievers with Cross-Lingual BrowseComp-Plus

Deep research agents are increasingly evaluated on their ability to search for evidence, reason over retrieved sources, and produce grounded answers. Existing browsing benchmarks, however, largely assume that the user's query and the supporting evidence are written in the same language, leaving open whether agentic search systems can operate when relevant evidence appears in another language. We introduce XBCP (Cross-lingual BrowseComp-Plus), a controlled benchmark that preserves the English question-and-answer space of BrowseComp-Plus but varies the languages of the supporting documents. XBCP instantiates two complementary settings: in the cross-lingual setting, each query is paired with evidence in a single assigned language. In the multilingual setting, the full evidence corpus is distributed equally and randomly across 12 languages spanning high-resource and low-resource regimes. We evaluate four deep research agents using sparse and dense multilingual retrievers, measuring answer accuracy, evidence recall, search behavior, calibration, citation fidelity, and oracle retrieval. Results reveal substantial degradation when evidence is translated. Even strong, dense retrievers lose evidence recall, and agents become less calibrated and cite evidence less reliably. Notably, accuracy remains lower even when all gold evidence is supplied directly. These findings suggest that cross-lingual deep research exposes both retrieval failures and an independent, agent-side difficulty in integrating language-mismatched evidence.

04.
arXiv (CS.AI) 2026-06-15

Think Fast: Estimating No-CoT Task-Completion Time Horizons of Frontier AI Models

arXiv:2606.07157v2 Announce Type: replace Abstract: Many efforts to ensure frontier AI models are safe rely on monitoring their chain-of-thought (CoT) reasoning. If models become able to perform sufficiently complex reasoning internally, without explicit thinking tokens, this would undermine such oversight. We measure how well frontier models reason without CoT across a suite of over 30,000 questions spanning 43 benchmarks in domains including math, coding, puzzles, causality, theory-of-mind, and strategic reasoning. To compare models against humans, we estimate the $50\%$-task-completion time horizon (TH): the human time required for tasks a model completes with $50\%$ success rate. We complement this with a $50\%$ reasoning token horizon: the minimum number of o3-mini reasoning tokens needed for tasks a model solves with $50\%$ success rate. We find that the no-CoT $50\%$ TH of frontier models has been doubling roughly every year over the past six years, with GPT-5.5's TH reaching over 3 minutes and reasoning token horizon exceeding 1,500 tokens. Our median estimates predict that frontier no-CoT THs could exceed 7 minutes by 2028, and 25 minutes by 2030, though these projections carry substantial uncertainty. We recommend frontier developers track this explicitly.

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

CASR: A Robust Cyclic Framework for Arbitrary Large-Scale Super-Resolution with Distribution Alignment and Self-Similarity Awareness

Arbitrary-Scale SR (ASISR) remains fundamentally limited by cross-scale distribution shift: once the inference scale leaves the training range, noise, blur, and artifacts accumulate sharply. We revisit this challenge from a cross-scale distribution transition perspective and propose CASR, a simple yet highly efficient cyclic SR framework that reformulates ultra-magnification as a sequence of in-distribution scale transitions. This design ensures stable inference at arbitrary scales while requiring only a single model. CASR tackles two major bottlenecks: distribution drift across iterations and patch-wise diffusion inconsistencies. The proposed SSAM module aligns structural distributions via superpixel aggregation, preventing error accumulation, while SARM module restores high-frequency textures by enforcing correlation-guided consistency and preserving self-similarity structure through correlation alignment. Despite using only a single model, our approach significantly reduces distribution drift, preserves long-range texture consistency, and achieves superior generalization even at extreme magnification.

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

MASCOT-Android: A Curated Dataset and Automated Collection Pipeline for Android Malware Source Code Specimens

arXiv:2606.16072v1 Announce Type: cross Abstract: Compared with binaries and decompiled code, malware source code more directly reflects the attackers' original intent. However, the scarcity of source code and the high cost of manual review make such datasets difficult to build and maintain. We propose MASCOT-Android, a curated dataset of Android malware source code and an automated collection framework for scalable malware source code discovery on GitHub. A key finding of our work is that repository-level documentation alone provides a strong signal for malware source code collection. Our model extracts character-level TF-IDF features from 8,772 malware and 25,747 benign README documents and trains a LinearSVC classifier to distinguish malware repositories. This README-only model achieves an accuracy of 96.28\% and an FPR of 1.06\% in local evaluation. In addition, the model outputs confidence scores, allowing users to adjust the decision threshold to balance FPR and coverage, which is practical in real-world malware source code collection.

07.
arXiv (CS.AI) 2026-06-15

MeEvo: Metacognitive Evolution Combined with Natural Evolution for Automatic Heuristic Design

arXiv:2606.14202v1 Announce Type: cross Abstract: Large Language Models (LLMs) have advanced Automatic Heuristic Design (AHD) by enabling heuristic generation through reasoning and code synthesis. Existing LLM-based AHD architectures mainly follow two paradigms: Natural Evolution, which uses crossover and mutation to explore heuristic programs, and Metacognitive Evolution, which refines reasoning through reflection. However, Natural Evolution discards reasoning traces, weakening knowledge inheritance and exploitation, while Metacognitive Evolution lacks population-level recombination, limiting exploration and increasing the risk of premature convergence. These limitations reduce search efficiency, stability, and solution quality on complex problems. To address this gap, we propose MeEvo, a dual-layer AHD framework that cyclically couples Natural Evolution and Metacognitive Evolution. Natural Evolution explores heuristic code while recording reasoning traces, fitness values, and errors into a shared history; Metacognitive Evolution then reflects on this history to generate improved heuristics that re-enter the parent pool for the next cycle. This design enables population-driven exploration and reflection-driven refinement to reinforce each other. Experiments on five optimization problems with two LLM backbones show that MeEvo achieves stronger and more stable performance than existing LLM-based AHD architectures, especially on complex constrained tasks.

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

QoS-Aware Token Scheduling and Private Data Valuation for Multi-Modal Agentic Networks

arXiv:2606.15573v1 Announce Type: new Abstract: In agentic systems, human-generated data records anchor the value of AI services. Yet cloud compute pipelines centralize processing on remote servers. Data centralization reduces personal data sovereignty and may potentially degrade the quality of service (QoS). Meanwhile, user contributions are diverse in quantity and quality: decentralized records can be biased, noisy, and heterogeneously distributed. To address the data challenge, we study fair token allocation and private data valuation for decentralized and resource-constrained agentic systems. Our approach embeds multi-modal representations in a shared semantic space and releases differentially private (DP) prototypes to preserve utility while reducing semantic leakage. With the DP guarantee, we design a fair token allocation scheme that rewards effective contributions and remains robust to data heterogeneity and AI resource scarcity. Extensive simulations demonstrate improved contribution-based fairness and QoS compared to standard benchmarks. The improved resistance to image reconstruction attacks indicates enhanced privacy for multi-modal personal data.

09.
arXiv (CS.CL) 2026-06-17

A Multifaceted Analysis of Social Biases in Large Language Models

Large language models (LLMs) have rapidly become indispensable tools for acquiring information and supporting human decision-making. However, ensuring that these models uphold fairness across varied contexts is critical to their safe and responsible deployment. In this study, we undertake a comprehensive examination of four widely adopted LLMs, probing their underlying biases and inclinations across the dimensions of politics, ideology, alliance, language, and gender. Through a series of carefully designed experiments, we investigate their political neutrality using news summarization, ideological biases through news stance classification, tendencies toward specific geopolitical alliances via United Nations voting patterns, language bias in the context of multilingual story completion, and gender-related affinities as revealed by responses to the World Values Survey. Results indicate that while the LLMs are aligned to be neutral and impartial, they still show biases and affinities of different types.

10.
arXiv (CS.CL) 2026-06-15

"I Didn't Make the Micro Decisions": Measuring, Inducing, and Exposing Goal-Level AI Contributions in Collaboration

As large language models (LLMs) increasingly shape how users form, refine, and extend their goals, attributing contributions in human-AI collaboration becomes critical for users calibrating their own reliance and for evaluators assessing AI-assisted work. Yet existing methods focus on final artifacts, missing the process through which goals themselves are jointly shaped. We introduce a goal-level attribution framework, CoTrace, that decomposes explicit goals into verifiable requirements and traces both direct contributions and indirect influences across dialogue turns. Applying CoTrace to 638 real-world collaboration logs, we find that while models account for only 11-26% of goal-shaping contribution, they contribute substantially more on introducing lower-level concrete requirements, and make various kinds of indirect contributions. Through controlled simulations, we show that interaction design choices significantly affect model goal-shaping behavior. In a user study, exposing participants to goal-level analyses shifts their perceived contributions by nearly 2 points on a 5-point scale, revealing systematic miscalibration in how users understand their own AI-assisted work.

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

GraphInfer-Bench: Benchmarking LLM's Inference Capability on Graphs

Graph analysis underlies many applications whose answers cannot be looked up in a single record or retrieved along a path: laundering rings, drug repurposing, user preference, and scientific theme are all inferred from a node together with its neighbourhood. We introduce GraphInfer-Bench, a benchmark for whether LLMs can perform this graph inference: producing an open-ended answer that no single node supports and no path retrieves. Existing graph-QA protocols cannot test this capability: algorithm simulation, node classification, single-node description, KG-QA, and GraphRAG all admit answers retrievable from one node or along a path. GraphInfer-Bench defines five tasks along Description (what a region is) and Comparison (how regions differ), each constructed so the ground truth lives in no single node. The release contains 42,000 samples across six real-world graphs, produced automatically and screened by a four-layer quality-control protocol. We evaluate four method families against the same tasks: graph-token alignment models, zero-shot frontier closed-source LLMs, Graph2Text supervised fine-tuning, and plain GNNs as a structural reference. No method family closes the gap. Graph-token alignment partially handles description tasks (relational, theme) but collapses on comparison tasks. Frontier LLMs lead on outlier detection and community partition among LLM-based methods but lag on masked-node prediction. Graph2Text SFT is the strongest LLM-based method on the description side yet falls behind frontier LLMs on comparison. Across every task, plain GNNs match or beat the strongest LLM-based row, with the largest margin on community detection. GraphInfer-Bench surfaces graph inference as an open capability gap rather than a property of any one architecture.

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

Stop the Sampler! Classifier-Based Adaptive Stopping for Sampling Kernels

arXiv:2606.16073v1 Announce Type: new Abstract: Sampling from complex, unnormalized probability densities is a fundamental challenge in Bayesian inference and probabilistic modeling. While Markov chain Monte Carlo (MCMC) methods provide asymptotic guarantees, they often suffer from slow mixing and high computational costs due to fixed or manually tuned trajectory lengths. In this work, we propose a novel framework that treats trajectory termination as a learnable component of the sampling dynamics. By framing MCMC within the theory of non-acyclic generative flow networks (GFlowNets), we train state-dependent neural classifiers to decide when a trajectory has reached a high-density region and should terminate. We theoretically establish the connection between optimal classifiers and the target density via detailed balance conditions and introduce a multilevel training scheme to facilitate exploration in complex geometries. Experimental results across various benchmark densities demonstrate that our approach significantly reduces average trajectory lengths while improving mode coverage and mixing compared to standard MCMC baselines.

13.
medRxiv (Medicine) 2026-06-22

National trends and operational drivers of vaccine wastage in Uganda, 2020-2025: a descriptive analysis of four tracer antigens

Background Vaccine wastage reduces immunisation efficiency, increases costs, and complicates supply forecasting. Uganda routinely monitors vaccine use, but national evidence comparing observed wastage with World Health Organization (WHO) and Uganda-specific planning thresholds has been limited. We described national and sub-national trends for four tracer antigens to inform supply-chain planning and forecasting. Methods We conducted a retrospective descriptive analysis of routinely reported immunisation data from Ugandas District Health Information Software 2, 2020-2025. We analysed Bacille Calmette-Guerin (BCG), measles-rubella (MR), oral polio vaccine (OPV), and diphtheria-tetanus-pertussis-containing vaccine (DPT). Vaccine wastage was calculated as the proportion of issued doses not administered. Annual wastage rates were summarised using medians, and temporal trends were assessed using the Mann-Kendall test. Observed wastage was compared with WHO thresholds: BCG[≤]50%, MR[≤]25%, OPV[≤]10%, DPT[≤]15%, and Ugandas planning thresholds: BCG[≤]70%, MR[≤]40%, OPV[≤]15%, DPT[≤]10%. Effective Vaccine Management reports were reviewed to summarise reported reasons for wastage. Results During 2020-2025, median national wastage was 40.6% for BCG, 25.9% for MR, 10.0% for OPV, and 9.2% for DPT. OPV wastage declined from 12.8% in 2020 to 8.0% in 2025, with a significant downward trend ({tau}b=-1.00; p=0.008). OPV and DPT wastage remained largely within their respective Uganda in-country thresholds ([≤]15% and [≤]10%) for most of the study period, while BCG generally remained below the WHO threshold ([≤]50%) and MR frequently exceeded the WHO threshold ([≤]25%) but remained within Uganda's planning threshold ([≤]40%) in most years. The proportion of districts exceeding both WHO and Uganda thresholds declined for OPV from 36.3% to 5.5% (p=0.024) and for DPT from 22.6% to 1.4% (p=0.013). Wastage was consistently higher in lower-level (Health Centre II and III) facilities, compared to hospitals. Among 50 service delivery points, reported reasons included low session attendance (66%), multi-dose vial policy non-compliance (28%), and vaccine expiry (12%). Conclusion Uganda achieved reductions in OPV wastage and district-level improvements in DPT wastage, while BCG and MR remained more variable and frequently had higher wastage. Strengthening adherence to the multi-dose vial policy and improving session planning at lower-level facilities could strengthen vaccine utilisation and forecasting.

14.
arXiv (CS.AI) 2026-06-12

Will AI Agents Free Us From Meaningless Work? A Human-Centered Analysis

arXiv:2606.12430v1 Announce Type: cross Abstract: Some claim that AI agents will free workers from the boring parts of their jobs, yet little is known about how workers themselves identify which tasks should be automated. Prior research focuses on occupations, overlooking that workers experience varying levels of meaning across tasks within the same role. We address this gap with a task-level analysis grounded in Graeber's theory of bullshit jobs. Using ratings from 202 workers on 171 workplace tasks, we (1) validate a five-item scale of perceived bullshitness, (2) show that perceived bullshitness strongly predicts desire for AI delegation, and (3) find that such tasks are also seen as requiring less human oversight. Together, these findings suggest that tasks perceived as bullshit are natural candidates for AI delegation, aligning worker preferences with perceived feasibility.

15.
arXiv (CS.AI) 2026-06-16

Artificial Intelligence Index Report 2026

arXiv:2606.15708v1 Announce Type: new Abstract: Welcome to the ninth edition of the AI Index report. As AI continues to advance rapidly, the question becomes whether the systems built around it can keep up. Governance frameworks, evaluation methods, education systems, and the data infrastructure needed to track AI's impact are struggling to match the pace of the technology itself. That gap between what AI can do and how prepared we are to manage it runs through every chapter of this year's report. New in this edition, the report tracks how AI is being tested more ambitiously across reasoning, safety, and real-world task execution, and why those measurements are increasingly difficult to rely on. It also features new estimates of generative AI's economic value alongside emerging evidence of its labor market effects, an analytical framework on AI sovereignty, and a science chapter developed in collaboration with Schmidt Sciences. For the first time, the report features standalone chapters on AI in science and AI in medicine, reflecting AI's growing impact across these two domains.

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

Understanding Sample Efficiency in Predictive Coding

arXiv:2605.11911v2 Announce Type: replace Abstract: Predictive Coding (PC) is an influential account of cortical learning. Much of recent work has focused on comparing PC to Backpropagation (BP) to find whether PC offers any advantages. Small scale experiments show that PC enables learning that is more sample efficient and effective in many contexts, though a thorough theoretical understanding of the phenomena remains elusive. To address this, we quantify the efficiency of learning in BP and PC through a metric called ``target alignment'', which measures how closely the change in the output of the network is aligned to the output prediction error. We then derive and empirically validate analytical expressions for target alignment in Deep Linear Networks. We show that learning in PC is more efficient than BP, which is especially pronounced in deep, narrow and pre-trained networks. We also derive exact conditions for guaranteed optimal target alignment in PC and validate our findings through experiments. We study full training trajectories of linear and non-linear models, and find the predicted benefits of PC persist in practice even when some assumptions are violated. Overall, this work provides a mechanistic understanding of the higher learning efficiency observed for PC over BP in previous works, and can guide how PC should be parametrised to learn most effectively.

17.
arXiv (CS.AI) 2026-06-18

X+Slides: Benchmarking Audience-Conditioned Slide Generation

arXiv:2606.19256v1 Announce Type: new Abstract: Automatically generating slide decks from source documents is an important application of large language models (LLMs). Existing benchmarks primarily assess slide completeness and technical depth, while overlooking the target audience as a critical real-world factor. For instance, specialists demand rigorous proofs, whereas decision-makers prioritize actionable conclusions. To bridge this gap, we introduce X+Slides, a benchmark specifically designed for audience-conditioned slide generation. Built on a diverse corpus spanning 113 topics and seven presentation scenes, X+Slides employs a dynamic evaluation framework constructed from 8,133 deduplicated, source-grounded probes. By assigning audience-specific utility weights to the same source-grounded probes, X+Slides reports four complementary metrics: Audience Coverage measures how much audience-essential information is conveyed, Domain-wise Coverage shows which information types are covered, Efficiency measures delivered utility per unit of attention cost, and Correctness verifies whether slide claims are supported by the source. Experiments on DeepPresenter, SlideTailor, and NotebookLM show that current systems can recover a substantial but still incomplete part of audience-essential information: at $\tau_A=0.7$, DeepPresenter reaches a best Audience Coverage of 0.714, SlideTailor reaches 0.594, and the NotebookLM ablation reaches 0.853 while showing clear grounding differences. These results indicate that visual quality and broad topic coverage should not be treated as evidence support without source-grounded evaluation.

18.
arXiv (quant-ph) 2026-06-16

Neural quantum states for entanglement depth certification from randomized Pauli measurements

arXiv:2512.13121v2 Announce Type: replace Abstract: Entanglement depth quantifies how many qubits share genuine multipartite entanglement, but certification typically relies on tailored witnesses or full tomography, both of which scale poorly with system size. We recast entanglement-depth and non-$k$-separability certification as likelihood-based model selection among neural quantum states whose architecture enforces a chosen entanglement constraint. A hierarchy of separable neural quantum states is trained on finite-shot local Pauli outcomes and compared against an unconstrained reference model trained on the same data. When all constrained models are statistically disfavored, the data certify entanglement beyond the imposed limit directly from measurement statistics, without reconstructing the density matrix. We validate the method on simulated six- and ten-qubit datasets targeting GHZ, Dicke, and Bell-pair states, and demonstrate robustness for mixed states under local noise. Finally, we discuss lightweight interpretability diagnostics derived from trained parameters that expose coarse entanglement patterns and qubit groupings directly from bitstring statistics.

19.
arXiv (CS.AI) 2026-06-17

SEAGym: An Evaluation Environment for Self-Evolving LLM Agents

arXiv:2606.17546v1 Announce Type: new Abstract: Self-evolving LLM-based agents improve mainly by changing their agent harness: the structured execution layer around a base model, including prompts, memory, tools, middleware, runtime state, and the model-tool interaction loop. Existing evaluations often reduce this process to isolated task scores or a single sequential curve, obscuring whether an update produces reusable improvement, overfits recent tasks, increases cost, or harms older behavior. We introduce SEAGym, an evaluation environment for measuring agent harness updates across training, validation, test, replay, and cost records. SEAGym turns Harbor-compatible benchmarks into dynamic self-evolution task sources with train batches, frozen update-validation, held-out ID and OOD transfer views, replay diagnostics, and saved snapshot and metric records. Instantiating SEAGym on Terminal-Bench 2.0 and HLE, we compare ACE, TF-GRPO, and AHE under a shared epoch/batch protocol. The results show that these evaluation views provide complementary signals about the evolution process: frequent updates may fail to improve held-out performance, useful intermediate snapshots may collapse later, and source diversity and model backend can affect harness reliability.

20.
arXiv (CS.LG) 2026-06-17

ReRAM-aware Model Finetuning addressing I-V Non-linearity and Retention Errors

arXiv:2606.17471v1 Announce Type: new Abstract: Traditional CPU, GPU, and NPU architectures are increasingly limited by the von Neumann bottleneck. While In-Memory Computing (IMC) using ReRAM crossbar arrays offers a high-density, energy-efficient alternative, its practical deployment is constrained through their non-idealities. Existing hardware-aware training frameworks often require training from scratch, which is computationally prohibitive for modern large-scale models. In this work, we propose a finetuning-based hardware-aware training algorithm that enables robust DNN deployment on ReRAM with minimal training overhead. Our approach mitigates I-V non-linearity by applying a range-shrunk sinh transformation and incorporates retention errors directly into a regularization loss during the finetuning process. We evaluate our framework across models and tasks such as image classification and question-answering (QA). Experimental results demonstrate that our method achieves similar accuracy on large-scale models like ResNet18 and DeiT-Tiny as the base model. In-case of ImageNet for MobileNetV3 families the technique has only less than 2% accuracy degradation. Further, applying the technique on the SQuAD v2 dataset results in only 1 point degradation of F-1 score.

22.
arXiv (CS.CV) 2026-06-16

Token-Level Entropy Reveals Demographic Disparities in Language Models

We ask whether demographic identity, signaled by a name alone, systematically reshapes the generative distribution of a language model. Measuring full-vocabulary Shannon entropy at temperature zero across six open-weight base models and 5,760 implicit sentence-completion prompts (e.g., "Tanisha walked into the office on a Monday morning and"), we find that Black-associated names produce higher first-token entropy than White-associated names across all six architectures - opposite to the output-level homogeneity bias documented under explicit demographic prompting (Lee et al., 2024) - and Black-associated names always produce greater entropy above identity-neutral baselines than White-associated names ($\Delta\Delta > 0$ in all six models). Women-associated names co-occur with lower first-token entropy (DL-pooled $\hat\beta = -0.041, p = .019$) and more homogeneous outputs ($\hat\alpha = +0.024, p < .001$) than men-associated names - a pattern convergent with homogeneity bias; race and gender effects are additive. Instruction tuning does not attenuate the race gap (matched-format DL-pooled $\hat{\beta}=+0.153$). Running the same templates with explicit group labels instead of names yields null race effects in 10 of 12 models where implicit probing is significant - establishing that probing methodology is a primary determinant of which distributional structure is recovered.

23.
arXiv (CS.CV) 2026-06-17

SPHINX: First Explain, Then Explore

Generating adversarial driving scenarios is critical for evaluating and improving autonomous vehicle decision-making systems in simulation. Recent approaches, such as ChatScene and LLM-Attacker, rely primarily on the prior knowledge of Large Language Models and Vision-Language Models to generate driving scenarios procedurally. We argue that adversarial scenes should be generated based on the failure diagnosis (e.g., indecisiveness, multi-frame inconsistency) of the driving policy to specifically address the policy's weaknesses instead of relying on prior assumptions. In this paper, we propose SPHINX, a closed-loop framework for adversarial scenario synthesis guided by a simple principle: first explain, then explore. Beyond blindly exploring the scenario space, SPHINX leverages explainable artificial intelligence methods to analyze the policy, identifying key visual concepts and their influence on policy outputs, and the uncertainty of the decisions. Given the interpretable evidence extracted from the policy's own decision process, we use a vision language model to rationalize and criticize failure modes of the current policy. These critics are then used to generate targeted adversarial scenarios for policy retraining and improvement. We demonstrate that SPHINX can highlight an interpretable account of policy failures while other adversarial scene generation cannot. Across the evaluated benchmarks and test suites, SPHINX can be applied to diverse state-of-the-art autonomous vehicle architectures and yields consistent robustness improvements over existing scenario-generation methods.

24.
arXiv (CS.CL) 2026-06-18

LLM Compression by Block Removal with Constrained Binary Optimization

In this paper, we formulate the compression of large language models (LLMs) by optimally deleting transformer blocks (``block removal'') as a constrained binary optimization (CBO) problem that can be mapped to a physical system (Ising glass), whose energies are a strong proxy for downstream model performance. This formulation enables an efficient ranking of a large number of candidate block-removal configurations yielding many high-quality, non-trivial solutions beyond those only removing consecutive regions. Our method performs strongly in the deep compression regime, such as for 50% compression of Llama-3.3-70B-Instruct, where we achieve an almost 23 percentage point increase on the MMLU benchmark compared to other state-of-the-art (SOTA) block-removal methods. For lighter compression, it performs on par with those methods across several benchmarks for Llama-3.1-8B-Instruct, Qwen3-14B (both before and after retraining), as well as Llama-3.3-70B-Instruct. The approach is computationally efficient and requires only forward and backward passes on a calibration dataset for a few active parameters. Additionally, we demonstrate that using good heuristic solvers for the CBO problem provides solutions that perform well on downstream tasks in negligible runtime when it is unfeasible to solve the problem exactly. The method can be readily applied to any architecture. We illustrate this generality on the recent NVIDIA-Nemotron-3-Nano-30B-A3B-FP8 model, which exhibits a highly inhomogeneous and challenging block structure, and where we outperform SOTA for AIME25 and GPQA when removing either 2 attention layers or 3 mixture-of-experts layers.

25.
arXiv (CS.CL) 2026-06-19

Proactive Conversational Assistant for a Procedural Manual Task based on Audio and IMU

Real-time conversational assistants for procedural manual tasks often depend on video input, which can be computationally expensive and compromise user privacy. For the first time, we propose a real-time conversational assistant that provides comprehensive guidance for procedural manual tasks using only lightweight privacy-preserving modalities such as audio and IMU inputs from a user's wearable device to understand the context. Using a furniture assembly task and a cooking task, we show how this assistant proactively communicates step-by-step instructions to a user performing a procedural task, and answers user questions. We illustrate the data generation method and the system design to achieve such an assistant. On observing that an off-the-shelf language model is a talkative assistant but is not always able to answer questions correctly, we demonstrate how finetuning the model improves its ability to limit unnecessary dialogues with a 50% increase in the precision, while also improving its ability to answer questions correctly, measured by a 150% increase in the recall of answers. We further describe how such an assistant is implemented on an edge device with no dependence on the cloud.