Computer Science (arXiv)

Real-world robots need to adapt their behavior beyond the envelope of their pre-trained policy. Policy finetuning or retraining are options, but they risk catastrophic forgetting, degrading the pretrained policy's base performance. To combat this, we introduce CLAE: Closed-Loop Affine Activation Editing, an inference-time framework for steering the behavior of a frozen policy by editing intermediate activations while keeping the base policy weights and downstream action head untouched. CLAE approaches behavior steering as a closed-loop problem whose outputs edit policy activations that adapt online to the robot state, environment, target behavior, and multi-robot context. It trains a sparse autoencoder over frozen-policy activations, selects behavior-relevant latent features via post-hoc probing, and learns a lightweight RL-based steering policy that applies state-dependent affine edits to selected latents during inference. We validate CLAE on a frozen multi-quadrotor navigation policy trained to perform a single task: navigating robots to a set of goal locations while avoiding obstacles. Through extensive simulations and physical tests, we show that while navigating to their goal positions, CLAE can 1. steer individual robot behavior by controlling each robot's velocity profile; 2. coordinate multirobot behavior by preserving a desired formation; and 3. produce entirely new behavior wherein robots are required to reduce their exposure to surveillance cameras in the environment.

Barcode scans, clear phone calls, reliable data storage, satellite communication, and large-scale quantum computation are all made possible by error correction. We present a handbook version of The Error Correction Zoo, a curated reference of methods for protecting classical or quantum information from errors during storage and transmission. The handbook includes descriptions of these error-correcting codes and a classification according to the symbols they use. It also catalogues relations among codes and related objects such as sphere packings, lattices, designs, groups, and classical and quantum phases of matter. The collection is intended both as a rigorous reference and as a practical aid for tracing the web of code relationships and uncovering new connections.

We study the undirected three-terminal reachability-preserving minimum edge cut problem. The input is an undirected graph $G=(V,E)$ with nonnegative edge costs, two protected terminals $s_1,s_2$, and a target terminal $t$. The goal is to remove a minimum-cost edge set so that $t$ is disconnected from the protected terminals while $s_1$ and $s_2$ remain connected. This problem captures a basic tension between separation and connectivity preservation. Prior work on connectivity-preserving cuts established polynomial-time solvability for some special cases, such as planar edge-cut instances, and strong hardness for node-cut variants, but a general-graph approximation guarantee for the undirected three-terminal edge-cut version does not appear to have been known. We give a polynomial-time $O(\sqrt n)$-approximation algorithm in this paper. This is the first known approximation algorithm for the problem

Understanding and predicting how social beliefs evolve in response to events -- from policy changes to scientific breakthroughs -- remains a fundamental challenge in social science. Given LLMs' commonsense knowledge and social intelligence, we ask: Can LLMs model the dynamics of social beliefs following social events? In this work, we introduce the concept of the Social World Model (SWM), a general framework designed to capture how social beliefs evolve in response to major events. SWM learns state-transition functions for social beliefs by mining temporal patterns in social data and optimizing the evidence lower bound, without the need for explicit human annotations linking events to belief shifts, or for expensive census data. To evaluate SWM, we introduce a benchmark, SWM-bench, derived from real-world prediction markets, specifically Kalshi and Polymarket. SWM-bench includes over 12k data points for social belief prediction tasks spanning diverse domains such as politics, finance, and cryptocurrency. Our experimental results show that SWM significantly outperforms time-series foundation models, achieving state-of-the-art results on Kalshi data and demonstrating competitive performance on Polymarket data, while offering interpretable insights into the underlying mechanisms of social belief dynamics.

Federated continual learning (FCL) must learn from distributed task streams under limited resources, such as communication, computation, memory, and label availability. Existing FCL methods often rely on repeated local optimization, replay, and full supervision. Analytic alternatives avoid iterative training and replay, but using high-dimensional random features to improve accuracy requires a second-order feature statistic, the Gram matrix, which has a quadratic communication cost in the random feature size $M$. We propose FedRAN, a resource-aware analytic FCL framework that replaces gradient-based updates with compact random feature statistics. Each client transmits a truncated-SVD summary of its Gram matrix, reducing the dominant second-order upload from quadratic to linear in $M$ for fixed rank. The server performs a two-level QR-SVD subspace merge, spatially across clients and temporally across tasks, and solves a ridge classifier in closed form. FedRAN further supports label scarcity through prototype-based pseudo-labeling. Across CIFAR-100, ImageNet-R, and VTAB datasets, FedRAN improves average accuracy by up to 4.8 percentage points over the strongest baseline, uses 30.6-121.8$\times$ less per-client communication than optimization-based FCL, and is 190.3$\times$ faster on average than gradient-based baselines; with only 20% labels, pseudo-labeling improves average accuracy by up to 6.61 points. These results show that FedRAN enables accurate and resource-efficient FCL under communication, computation, and label constraints. The source code is available at https://github.com/JebacyrilArockiaraj/Fed-RAN-SSL.

Correcting handwritten exams by hand is time-consuming and error-prone, particularly for large cohorts, while fully digital exams tend to force a didactic narrowing towards closed question formats. A practical middle ground keeps paper-based, problem-oriented tasks but records the assessment-relevant answers as single capital letters in a table that a machine can read. The open question is whether this reading can be made accurate and, above all, fair enough for unsupervised grading. Earlier automated approaches reached only about 88%--91% recognition -- too low -- and failed on the cases that matter most: answers placed outside the cell, crossed out, or written in cursive. We show that general-purpose vision-language foundation models (VLMs), which interpret the page rather than match pixel templates, close this gap. On a benchmark of 61 anonymised exams (3141 answer positions) the best model reaches 98.4% accuracy, well above the previous baseline. Crucially, we centre the evaluation on fairness: we distinguish false negatives (a correct answer marked wrong, which disadvantages the student) from false positives, and a lightweight prompt that supplies the reference solution as context lowers the false-negative rate to 0.58%. Under an exemplary grading scheme only three of the 61 exams would be graded worse, all caught by a student self-review step. Fully automated, fairness-aware exam grading at scale is therefore defensible; we release the anonymised benchmark to support reproducibility.

Software engineering teams increasingly depend on GitHub issue threads to coordinate work, report bugs, and negotiate technical decisions, yet most repository health tools focus on code metrics and ignore the conversational dynamics that drive or stall development. This paper presents SentTrack, a dual-lens framework for detecting socio-technical bottlenecks from GitHub issue discussions. Applied to the AvaloniaUI open-source repository across approximately 9,000 issue threads, the framework addresses three questions: how to automate workflow-inefficiency detection from real-time conversational data, whether sentiment signals can surface risk earlier than traditional label-based methods, and how to isolate human narrative from machine-generated noise in mixed-media issue text. SentTrack combines two complementary pipelines. A horizontal pipeline translates raw issue reports into clean summaries using a large language model, extracts mid-level concern phrases, and clusters them through UMAP and HDBSCAN, producing 613 semantic clusters from the first 3,608 issues processed. A vertical pipeline applies the ABCDE collaborative interaction framework to classify each comment and infer thread-level outcomes. Across the full corpus, 49\% of threads ended in stagnation and only 13\% reached resolution, with the resolution gap identified as the dominant bottleneck signal. A weighted scoring engine that combines negativity, stagnation, resolution gap, and thread length gives maintainers an interpretable prioritization tool for high-friction discussions before they stall development.

In this paper, we study Mahalanobis-guided latent out-of-distribution (OOD) detection for test-time RL controller switching in nonlinear time-varying systems. RL controllers can quickly control high-dimensional systems within the training distribution, but their performance can degrade when time-varying dynamics produce unseen observations. We consider a combined ES--DRL controller, where RL provides fast in-distribution actions and bounded extremum seeking (ES) provides robust model-independent control under OOD operation. The key challenge is deciding when to switch. We train a variational autoencoder (VAE) on in-distribution beam-profile observations and use Mahalanobis distance in the VAE latent space to detect OOD beam profiles at test time. This OOD decision sets a binary switch that selects either the RL controller or the ES controller. We evaluate the approach in safety-critical particle accelerator control. In this setting, spatial magnet motion creates OOD beam profiles that were not seen during RL training. Visualization of the VAE latent space shows that the proposed method identifies this OOD scenario and provides an interpretable signal for switching between RL and ES in the combined controller.

Prior-fitted networks (PFNs) are a promising class of tabular foundation models that perform in-context learning, whereby the entire labelled training set is supplied as context, and predictions for test queries are produced in a single forward pass. However, the quadratically scaling self-attention mechanism in many PFN architectures makes inference prohibitive for very large training datasets. We propose CRUMB (Clustered Retrieval Using Minimised-MMD Batching), a three-stage inference wrapper that (i) clusters the test queries, (ii) selects a small, distributionally matched training subset for each cluster by greedily minimising the maximum mean discrepancy (MMD), and (iii) runs exact PFN inference on each reduced-context batch. CRUMB is architecture-agnostic and requires no retraining. On the 51-dataset TabArena benchmark, evaluated across three PFN architectures (TabPFNv2, TabICLv1, TabICLv2), we show that CRUMB outperforms similar state-of-the-art context selection strategies. We also show that CRUMB is resilient to covariate drift, as the MMD-minimisation step naturally helps align the training context distribution to match the current test batch distributions.

The expansion of the digital domain has resulted in a substantial increase in digital communication, with email emerging as one of the most prominent channels. The proliferation of email communication is apparent in both professional and personal contexts, thereby creating numerous vulnerabilities for malicious actors to exploit. Spam emails, a form of unsolicited correspondence often bearing malicious intent towards recipients, have been an ongoing challenge for email users since the inception of email technology, and this problem has been exacerbated by the growth of the digital landscape. Email spam filters are integral components of email clients, engineered to identify potentially harmful messages and alert users to their malicious content. Phishing, frequently the initial phase of malware-based attacks, is evolving rapidly, with malware becoming increasingly sophisticated over time. A widely adopted approach for detecting malicious activity within malware and spam domains is the application of machine learning. Our aim is to assess the impact of the evolution within the spam email domain on these machine learning-based detection systems and to explore strategies for mitigating associated performance degradation.

Large Language Models (LLMs) have achieved strong performance across natural language processing tasks, yet reliable reasoning remains an open challenge. Although modern LLMs show progress in structured inference, multi-step problem solving, and contextual understanding, their reasoning behavior is often inconsistent and sensitive to prompting strategies, task design, and model scale. This survey provides a systematic analysis of more than 300 recent papers from arXiv, Semantic Scholar, Google Scholar, Papers with Code, and the ACL Anthology to examine how reasoning capabilities emerge in LLMs and where they fail. We make three main contributions. First, we introduce a structured taxonomy of LLM reasoning research, covering Chain-of-Thought reasoning, multi-hop reasoning, mathematical reasoning, common sense reasoning, visual and temporal reasoning, code and algorithmic reasoning, retrieval-augmented reasoning, tool-augmented and agentic reasoning, and reinforcement learning-based reasoning. Second, we analyze methodological trends across these paradigms, including prompting methods, model architectures, training objectives, reward modeling, and evaluation benchmarks. Third, we synthesize recurring limitations and failure modes, such as reasoning hallucinations, brittle multi-step inference, weak causal abstraction, and poor cross-domain generalization. By organizing a rapidly expanding literature, this survey offers a unified view of the current capabilities and limitations of reasoning in LLMs. We also identify emerging research directions, including meta-reasoning, self-evolving reasoning frameworks, multimodal reasoning, and socially grounded reasoning. Overall, this work aims to serve as a reference for developing more robust, interpretable, and generalizable reasoning systems in future language models.

We study the task of density estimation, where we hope to accurately estimate a probability density from $n$ samples. A textbook method for density estimation in total variation distance is the minimum-distance estimator approach, where we conclude both the algorithm and the analysis merely from bounding the VC dimension of a particular concept class (the so-called Yatracos class). While this technique has originally yielded sharp guarantees primarily for total variation distance, in this work we extend the minimum-distance estimator approach for learning within Hellinger distance. Our main observation is that we may produce an analogous recipe for Hellinger (where we only require bounding the VC dimension of a related concept class) by drawing connections to recent results yielding reverse data processing inequalities. This recipe is flexible enough to accommodate fast algorithms originally designed for total variation distance; by modifying the approach of Acharya et al. (2017) we conclude the first near-linear time algorithm for learning classes including univariate mixtures of log-concave densities and mixtures of Gaussians (with arbitrary variances), with near-optimal sample complexity.

Entanglement-assisted (EA) quantum QC-LDPC codes offer strong error-correction capabilities with structured parity-check matrices, but their practical use depends on efficient encoder circuits and the availability of pre-shared Bell pairs (ebits). In all encoder implementations based on the stabilizer formalism, the dominant contribution to this complexity comes from the use of controlled gates. In this paper, we adopt the Sharma-Kumar-Garani (SKG) encoder construction. We formulate the encoder optimization as a search over GF(2) row operations that decompose the binary matrix derived from its CNOT sub-sequence. We solve this problem using a beam search algorithm guided by a Hamming-distance heuristic. For the tested EA quantum QC-LDPC code families, the proposed method achieves CNOT-count reductions of 7.3-34.0% relative to the SKG baseline encoder. The optimized circuits also yield lower CNOT counts than Patel-Markov-Hayes synthesis on all tested instances and are verified by stabilizer-tableau simulation. These results show that substantial encoder simplification is possible for structured EA QC-LDPC codes.

Point cloud semantic segmentation requires architectures that capture both fine-grained local geometry and broad global scene structure. Transformer-based networks have demonstrated strong performance by focusing on detailed local feature aggregation; however, global context is conveyed primarily through skip connections across encoder-decoder stages, which we argue is insufficient for full scene understanding. We hypothesize that augmenting skip connections with a learnable global feature extraction module allows the network to acquire scene-level knowledge before descending into local detail, leading to richer and more contextually grounded representations. To this end, we propose Point Transformer with Wavelet Neural Operato (PT-WNO), which integrates a shared Wavelet Neural Operator (WNO) branch alongside the skip connections of a point cloud transformer backbone. At each encoder-decoder transition, point features are projected onto a dense 3D volumetric grid where the WNO captures multi-scale global spectral context through learnable wavelet decomposition and reconstruction. These global features are fused back into the network via lightweight adapters, complementing rather than replacing the existing skip connections. Experiments on four large-scale 3D point cloud benchmarks demonstrate the effectiveness of PT-WNO. On S3DIS (Area 5), PT-WNO achieves 71.59% mIoU, outperforming the Point Transformer v3 (PTv3) baseline by +1.03 points. On DALES it achieves 81.05% mIoU (+1.47 over the baseline). On ScanNet~v2, PT-WNO obtains 76.19% mIoU, remaining competitive with the baseline (76.36%).

Robotic table tennis is a representative benchmark for high-speed, closed-loop robotic control in dynamic environments, where accurate and fast prediction of ball states is critical for reliable planning and control. Physics-based approaches rely heavily on accurate parameter identification and precise initial state, while learning-based methods often struggle to capture long-range temporal dependencies and are typically trained on limited or simulated data. We propose a transformer-based framework for table tennis ball state prediction that leverages attention mechanisms to model long-range temporal correlations directly from historical observations, without relying on explicit flight or bounce models. To support robust learning and generalization, we collected a large-scale real-world dataset from players of varying skill levels and diverse ball cannon configurations. The combination of a high-capacity transformer architecture and extensive real-world data enables accurate long-horizon forecasting. Building on this capability, we introduce a plug-and-play sim-to-real transfer strategy, Swap Predictor at Deployment (SPAD), which replaces the physics-based simulator used during training with the proposed real-world-trained predictor at deployment, improving the sim-to-real transferability of the policy without requiring retraining. We demonstrate that this simple substitution effectively narrows the sim-to-real gap while preserving the efficiency and scalability of simulation-based training.

Accurate loss reserving is foundational to insurer solvency, yet accelerating climate driven catastrophes systematically violate the stability assumptions on which traditional actuarial methods depend. This white paper presents a research program testing whether Long Short Term Memory (LSTM) neural networks can detect and adapt to these structural breaks faster and more accurately than Chain Ladder, Bornhuetter Ferguson, and Cape Cod methods. Using 15 plus years of regulatory development triangle data from Florida and Louisiana, enriched with NOAA hurricane intensity indices and sea surface temperatures, we hypothesize a targeted improvement of 15, 20% in reserve accuracy for catastrophe exposed years, a threshold grounded both in the prior neural network reserving literature and in the formal convergence results developed here. Beyond empirical validation, we develop a theoretical framework grounding LSTM structural break detection in probabilistic terms, providing formal performance guarantees that compensate for the limited number of catastrophe events in the test period. We document the research design, methodology, expected contributions, and a candid assessment of limitations.

Safety assurance cases provide structured justifications that safety-critical systems meet their safety requirements. Recently, the notion of defeaters has emerged as a rigorous means of challenging the validity of safety arguments. Examples of defeaters might include overly strict claims, unreliable evidence, or reasoning gaps. However, defeaters remain ad hoc, lack structured support for critical reflection, are inconsistently described, are difficult to review, and lack documentation standards. To address this, we propose Defeater Cards, a new structured documentation artifact for systematically characterizing, reasoning about, and managing defeaters in safety cases. Drawing on a literature survey and thematic analysis, we identify documentation criteria that inform the card's structure, based on the 5W1H framework. Defeater Cards are designed to support informed analysis and evolution, improve traceability and auditability, and enable the reuse of defeater knowledge across systems and product variants. We demonstrate their applicability through two cross-domain case studies, showing how they expose hidden assumptions, surface reasoning gaps, and support ongoing safety assurance case evolution. To support adoption and community reuse, we also release an open-source repository of defeater cards as a baseline upon which researchers and practitioners can build and describe lessons learned.

Large Language Models are highly sensitive to prompt formulation, necessitating automatic prompt optimization to unlock their full potential. While evolutionary algorithms have emerged as the dominant paradigm, they suffer from a critical bottleneck: data efficiency. Current methods treat the development dataset as a static benchmark, wasting significant compute budget on uninformative data. In this work, we introduce APEX (Automatic Prompt Engineering eXpert), a novel framework that optimizes the data usage alongside the prompt search. APEX dynamically stratifies the dataset into Easy, Hard, and Mixed tiers based on the optimization lineage. By prioritizing the Mixed tier, which identifies the data where the LLM has mixed performance, we identify two high-leverage subsets: the addressable frontier for generating informative mutations and the rank-sensitive frontier for distinguishing candidate quality. We evaluate APEX across three diverse benchmarks: IFBench, SimpleQA Verified, and FACTS Grounding. Under a fixed budget of 5,000 evaluation calls, due to its data efficiency, APEX outperforms the initial prompt by an average of 11.2% on Gemini 2.5 Flash and 6.8% on Gemma 3 27B, demonstrating that a data-centric approach is key to efficient and effective prompt optimization.

Behavioral biometrics offer a promising approach for continuous authentication, but their fairness across demographic groups remains largely unexplored. This paper investigates gender bias in swipe-based authentication using the BBMAS (117 users) and ANTAL (71 users) datasets and evaluates XGBoost and DenseNet classifiers through False Acceptance Rate (FAR) and False Rejection Rate (FRR). XGBoost achieved authentication accuracies of 92% and 94% on the BBMAS and ANTAL datasets, respectively, while statistical tests (Kolmogorov-Smirnov, Mann-Whitney, and Wasserstein permutation) found no significant gender differences in authentication error rates across almost all experimental settings. These findings suggest that swipe-based authentication can achieve high accuracy while maintaining comparable performance for male and female users, supporting its potential as a fair and reliable behavioral biometric modality.

The deployment of LLM-based agents in scientific analysis raises opposing concerns: that agents may reduce methodological diversity, or that they may amplify the analytic flexibility through which researchers reach motivated conclusions. We argue these worries target two empirically separable layers: a design layer of methodological choices, and a verdict layer in which a decision rule maps estimates to a substantive claim. We test both by running 20 independent executions of Claude Code and Codex on a prominent immigration and social-policy against a many-analysts human baseline. At the design layer, Codex matches human methodological diversity and Claude Code produces nearly three times as many specifications; both agents' effect estimates remain broadly aligned with the human consensus, and no agent model exactly matches any human model. A prompt-induced anti-immigration researcher prior reorganizes each agent's methodological decisions but, unlike for biased human analysts in the same data, does not shift aggregate estimates or final verdicts; nor do agents reroute along the methodological axes humans use to bias their estimates. At the verdict layer, an explicit confirmatory prompt flips Claude Code's verdicts from 10% to 90% support while leaving its coefficient distribution essentially unchanged, operating through rule omission rather than rule softening. AI agents can rival or exceed human methodological diversity at the design layer while remaining vulnerable at the verdict layer. In our setting, the locus of AI bias is not estimation but interpretation.

Finite geometry (FG) codes combine the algebraic properties of classical block codes with the iterative belief propagation (BP) decoding ability of low-density parity-check~(LDPC) codes. However, exploiting both advantages in practice is hindered by the fact that the standard incidence matrix between $(μ+1)$-flats and points is dense and contains many short cycles for any flat dimension $μ\geq 1$. In this work, we propose to sparsify the decoding matrix based on pencil selection, formulated as a constant-dimension subspace packing problem and solved explicitly using lifted Gabidulin codes. For both affine and projective geometries, sparse parity-check matrices are constructed and verified for FG codes of lengths up to $1024$. Simulations on four FG codes show no visible error floor and around $0.5$~dB gain over corresponding 5G LDPC codes at a block error rate of $10^{-7}$.

Recently, masked skeleton reconstruction models have emerged as strong action representation learners, driving significant progress in self-supervised skeleton-based action recognition. However, existing state-of-the-art methods must predict an exceedingly large number of spatiotemporal patches, significantly prolonging training time. Besides, by treating all spatiotemporal regions equally during reconstruction, these models are distracted from learning the critical motion patterns that underlie action semantics. To address these challenges, we propose Adaptive Masked Reconstruction (AMR), a faster and stronger pre-training framework. We first decouple the decoder from the encoder, enabling flexible prediction of larger spatiotemporal patches and dramatically reducing reconstruction complexity. Given that larger patches contain more complex information, which is challenging to predict and consequently degrades performance, we accordingly introduce an adaptive guidance module. This module identifies regions of high motion informativeness, guiding the model to focus on the most discriminative parts of each patch and alleviating reconstruction difficulty. Experiments on NTU RGB+D 60, NTU RGB+D 120, and PKU-MMD datasets demonstrate that AMR not only accelerates pre-training substantially but also improves downstream recognition accuracy, surpassing current state-of-the-art approaches.

We study the query complexity of Boolean functions $f: \{0, 1\}^n \rightarrow \{0, 1\}$ in the noisy query model introduced by Feige, Raghavan, Peleg and Upfal [SICOMP 1994]. In this model, an algorithm can adaptively query the bits of an input vector, but each query result is independently flipped with constant probability $p \in (0, 1/2)$; repeated queries are allowed. The noisy query complexity $\mathsf{N}_p(f)$ of a function $f$ is defined as the minimum expected number of queries needed to compute $f(x)$ with error probability at most $1/3$, for the worst case input $x$. We prove a general lower bound on $\mathsf{N}_p(f)$ based on degree statistics of certain subgraphs of the Boolean hypercube. This is the first general lower bound beyond those implied by the simple observation that $\mathsf{N}_p(f)$ is lower bounded by the randomized query complexity. We show that this recovers (up to a constant factor) most previously known lower bounds on the noisy query complexity of Boolean functions, providing a unified framework for understanding these results and simplifying the proofs in several cases. Furthermore, this resolves in the affirmative an open problem of Gu, Li and Xu [COLT 2025] that $\mathsf{N}_p(f) = Ω(\mathsf{I}(f) \log \mathsf{I}(f))$, where $\mathsf{I}(f)$ denotes the total influence of $f$. We also apply our general lower bound to obtain tight bounds on the noisy query complexity for several new functions.

Recent anecdotal evidence suggests that AI coding agents can reproduce published findings when provided with original data and code; yet systematic evaluation across social sciences remains limited. Existing evaluation benchmarks are insufficient, either small or conflate agent performance with problems in the reproduction materials themselves, such as code that fails to execute correctly. Here we introduce SocSci-Repro-Bench, a benchmark of 221 tasks spanning four disciplines and 13 substantive domains, constructed from studies whose results are either fully reproducible with available materials or demonstrably non-reproducible due to missing data, allowing us to isolate agents' reproduction capacity. Evaluating two frontier coding agents, Claude Code and Codex, we find that both can reproduce a large share of social science findings, with Claude Code substantially outperforming Codex. These reproduction rates considerably exceed those previously reported for general-purpose LLM-based agents on comparable reproducibility benchmarks. Both agents also perform strongly on a reasoning task requiring identification of underlying research questions, and additional analyses suggest that results are not primarily driven by memorization. Providing the original paper PDF alongside replication materials modestly improves performance but introduces bias on tasks where reproduction is impossible. We also show that agents can be nudged toward confirmatory specification search through subtle prompt framing. Together, these findings suggest that at least some frontier coding agents can serve as reliable executors of computational workflows while underscoring the need for careful benchmarking and prompt design as AI systems assume larger roles in scientific production.

This research introduces a framework for incorporating Concept Bottleneck Models (CBMs) into 3D generative architectures to address the inherent 'semantic gap' in deep geometric learning. As deep models become central to 3D content creation, explainability shifts from a peripheral feature to a fundamental requirement for trust and accountability in safety-critical domains such as healthcare and manufacturing. CBMs provide an intrinsic interpretability solution by constraining latent representations to align with human-defined concepts, yet their application to unstructured 3D data remains largely unexplored. We design, implement, and validate a formal 3D-CBM architecture that maps raw geometric inputs, including point clouds and meshes, into a multi-tiered taxonomy of interpretable primitives and functional attributes. The framework further identifies strategic datasets, such as PartNet and ShapeNet, specialized for concept-based supervision. Experimental results from a 3D part-manipulation proof-of-concept experiment demonstrate the framework's efficacy, achieving a concept prediction accuracy of 88.8\% and a Chamfer Distance of 0.0115. Critically, the model enables precise test-time intervention, allowing for the interactive correction of structural errors. This work establishes a foundation for semantically-steerable 3D generation and invites further exploration into collaborative human-in-the-loop design systems.