paperlist.ai

AI conference papers

AI conference paper library

Search papers from 30+ top conferences, including NeurIPS, ICLR, ICML, CVPR, ACL and AAAI. Filter by keyword, then open any paper for its abstract, authors and links.

arXiv

Large-Small Model Collaboration for Enhancing Edge-Deployed Small Models

Peigen Liu, Yijiang Fan, Zixuan Xu, Yuren Mao, Longbin Lai, Ying Zhang

Edge devices host domain-specific small language models (SLMs) with limited resources, while private clouds offer larger LLMs. We propose G-Boost, an adaptive edge-cloud framework that improves a deployed SLM's task performance without parameter updates. It formulates reasoning as a tree search, choosing at each step between SLM-only inference and SLM-LLM logit fusion---which transfers domain knowledge from the SLM's adapted version to the cloud LLM without exposing private data. A process reward model guides Monte Carlo tree search to select beneficial collaboration steps dynamically. The edge runs the SLM and search controller; the cloud hosts the LLM and reward model, exchanging only current context. Evaluated on GSM8K and MATH-500 with Qwen2.5 and LLaMA2, G-Boost outperforms the SLM alone, static fusion, and fine-tuned baselines, gaining up to 8.6 and 10.7 percentage points over MCTS and Proxy-Tuning, respectively. Results confirm that step-level, reward-guided dynamic collaboration enhances reasoning and domain utilization for deployed edge SLMs.

arXiv

MathDebugger: Detecting and Diagnosing Errors in Synthetic Mathematical Data

Hao Liang, Meiyi Qiang, Yuying Li, Zefeng He, Xiaochen Ma, Ruichuan An, Yongzhen Guo, Zhengzhou Zhu, Bin Cui, Wentao Zhang

Synthetic mathematical data has become an important resource for scaling the reasoning capabilities of large language models, yet errors in generated questions and solutions can substantially undermine its value. We introduce MathDebugger, a type-aware benchmark for evaluating whether models can detect and diagnose errors in synthetic mathematical data. MathDebugger contains 2,000 correct questions, 2,000 erroneous questions balanced across four error types, and 2,000 annotated solutions, including 610 erroneous solutions spanning three error types. Each instance is manually verified and labeled for correctness, with erroneous instances further assigned a fine-grained error category. Human annotation achieves substantial to near-perfect agreement, with per-type Fleiss kappa ranging from 0.69 to 0.91. We evaluate 14 representative large language models and three process reward models. The results show that even strong reasoning models remain far from saturating MathDebugger, particularly when identifying fine-grained error types. We also uncover a consistent solving-verification gap: models specialized for mathematical reasoning or long-form reasoning do not necessarily outperform their general-purpose counterparts when auditing mathematical data. Finally, we show that explicit error-type information provides actionable guidance for correcting erroneous questions and solutions, yielding consistent and statistically significant improvements. MathDebugger provides a practical benchmark for developing more reliable mathematical data synthesis and quality-control pipelines.

arXiv

Unforgettable Generalization in Language Models

Eric Zhang, Leshem Choshen, Jacob Andreas

When language models (LMs) are trained to forget (or "unlearn'') a skill, how precisely does their behavior change? We study the behavior of transformer LMs in which tasks have been forgotten via fine-tuning on randomized labels. Such LMs learn to generate near-random predictions for individual examples in the "training'' set used for forgetting. Across tasks, however, LMs exhibit extreme variability in whether LM predictions change on examples outside the training set. In some tasks (like entailment classification), forgetting generalizes robustly, and causes models to produce uninformative predictions on new task instances; in other tasks (like physical commonsense reasoning and scientific question answering) forgetting affects only the training examples, and models continue to perform the "forgotten'' task accurately even for examples very similar to those that appeared in the training set. Dataset difficulty is not predictive of whether a behavior can be forgotten; instead, generalization in forgetting is (weakly) predicted by the confidence of LMs' initial task predictions and the variability of LM representations of training data, with low confidence and low variability both associated with greater generalization. Perhaps most surprisingly, random-label forgetting appears to be somewhat insensitive to the contents of the training set: for example, models trained on science questions with random labels continue to answer other science questions accurately, but begin to produce random labels on entailment classification tasks. Finally, we show that even generalizable forgetting is shallow: linear probes trained on LMs' representations can still perform tasks reliably after forgetting. Our results highlight the difficulty and unpredictability of performing targeted skill removal from models via fine-tuning.

arXiv

Offline Ambient-Controlled Latent Diffusion: Architecture, Telemetry, and On-Device Evaluation

Lech Kalinowski, Artur Morys-Magiera, Piotr Miłkowski

Most mobile image-generation applications are thin clients over cloud services, leaving outputs hard to audit. We present an Android latent-diffusion application that runs entirely on-device and is driven by the ambient-light sensor rather than a text prompt, keeping generation, telemetry, and storage local. The contribution is not a new diffusion method but the surrounding measurement workflow: each output is bound to the sensor reading, runtime path, and seed that produced it, giving a per-artifact audit trail for offline analysis. On a single Samsung foldable, one fixed capture of 373 artifacts shows the controller's log-lux input positively associated with output luminance (Pearson $r=0.532$, 95\% CI $[0.455, 0.601]$), confirming the ambient dependency survives denoising and VAE decoding, while the latent UNet/VAE pipeline runs at 552--1334\,ms mean latency across three quality tiers under the Android Neural Networks API (NNAPI).

arXiv

Phase-Aware CNN for Real-Time 5G/6G Channel Estimation with Hardware-in-the-loop Validation

Javad Zolfaghari-Bengar, Rakibul Rony, Elisa Gomez-de-Lope, Alejandro Villena-Rodriguez, Abhinav Mahadevan, Nicolas Kourtellis

In 5G/6G wireless systems, accurate and timely channel estimation is critical to ensure reliable communication under complex, fast-changing radio conditions. This work focuses on pilot-based channel estimation using deep learning to reconstruct both magnitude and phase across the full subcarrier grid, with particular emphasis on evaluation using emulated data collected from an end-to-end O-RAN testbed. The testbed includes hardware in the loop and controlled channel emulation to better reflect deployment conditions beyond pure software simulation. It addresses major limitations in classical estimators such as LS and MMSE, as well as deep learning-based approaches that struggle with phase prediction due to discontinuities at $\pm π$, poor generalization to different UE and antenna configurations, and computational inefficiency for real-time deployment. The proposed system combines a phase-aware input encoding using sine and cosine representations with a lightweight Convolutional Neural Network (CNN) architecture. This design achieves high accuracy, stable phase reconstruction, strong generalization across testbed-derived datasets, and real-time inference suitable for edge devices.

arXiv

Local verification cannot detect non-transportability: a cohomological theory of context preservation in agentic reasoning

Suyash Mishra

Agentic AI systems routinely transport conclusions across biological, clinical and financial contexts, and the emerging safeguard is local verification: checking at each step that the entity is representable in the chosen tool, that parameters are compatible, and that outputs cohere with the plan. We prove this class of safeguard is structurally incomplete. Modelling a covering of context space by its nerve and evidence by a real-valued 1-cochain, an agent chaining evidence performs path integration: its conclusion is path-independent if and only if the cochain is exact, and disagreement between valid reasoning paths is exactly the holonomy of a first Cech cohomology class. Hodge decomposition partitions evidence conflict into a gradient part (calibration), a curl part (local inconsistency, visible at triple overlaps) and a harmonic part. Our central result is that no family of simplex-supported consistency checks can distinguish omega from omega+h for harmonic h, which nonetheless generates non-zero disagreement between valid paths; detection requires a statistic on a cycle basis. The resulting procedure, Ksetra, estimates by coboundary projection and gates abstention on the harmonic component, which we give a mechanism: it arises from effect modification combined with overlap-specific population composition, and vanishes to machine precision when effect modification is absent. The degrees of freedom of an evidence network partition into calibration, coherence and transport, yielding an exact F-test for the existence of a global claim; we quantify its distortion under unequal precision and supply the precision-whitened form that restores exactness. Foreign exchange, where the arbitrage-free null makes the cochain exactly a coboundary, serves as a calibration bench: the test is correctly sized, fires on loop arbitrage, and ignores triangular arbitrage.

arXiv

AgonAlpha: Autonomous Alpha Discovery via Prompt Economy and Scalable Agentic Search

Weicheng Ye, Youran Sun, Xingyu Ren, Shunyao Yu, Chugang Yi, Haizhao Yang

Language models can propose many plausible trading factors, but an autonomous research system must also allocate its evaluation budget, verify its own evidence, and preserve how each candidate was produced. We present AgonAlpha, an architecture that searches over frozen research artifacts---hypotheses, executable expressions, platform evidence, rationales, and review status---rather than formulas alone. To our knowledge, AgonAlpha is the first alpha-mining system to combine verified artifact search, a fresh-context adversarial reviewer with re-execution and veto authority, and pending-aware parallel budget allocation, together with a complete public evidence trail. Independent deployments on WorldQuant BRAIN produced SPECTACULAR-grade alphas across five users and six model backends, with Fitness reaching 9.50 and Sharpe reaching 3.48, while retaining prompt-to-expression provenance for every submission.

arXiv

"YES! YES! I absolutely love this insight!" Affirmative Narration as Interactional Strategy in Dialogues with LLM Chatbots

Hanna-Riikka Roine, Anne Sigrid Refsum, Jill Walker Rettberg

This article analyses narrative mechanisms that are common in dialogues with LLM chatbots. In combination, these mechanisms produce an interactional strategy for maximising user engagement, which we call affirmative narration. Affirmative narration serves to convince users of the chatbot's utility. We analyse three narrative mechanisms that support affirmative narration in human-LLM dialogues: firstly, guiding the user to view the chatbot as an intelligent and reliable character; secondly, activating masterplots, culturally significant and recurring story templates; and thirdly, using characters and masterplots not only to affirm, but also to isolate the user. The case studies range from a journalist's unsettling chatbot experiment to cases where users have experienced delusions or even died by suicide after lengthy interactions with a chatbot. The analyses illustrate the worrying sides of affirmative narration, and the article thus concludes with a discussion of LLMs as a genre of fictional narrative media that requires a new type of literacy.

arXiv

Variable Selection in the Context of AI Fairness

Ivan Luciano Danesi, Chiara Frigerio, Fabio Maccaferri, Giorgio Alessandro Motta, Pietro Zecca

Fairness in AI systems has become more important with recent regulatory demands, such as the EU AI Act. Traditional approaches often do not take into account philosophical ethics and social awareness. Variable selection processes, in particular, can introduce implicit bias, affecting equity across different subgroups. We discuss a mathematical approach that evaluates fairness in AI, aligning mathematical methodologies with ethical considerations and regulatory requirements. Our aim is to advocate for interdisciplinary collaboration to address fairness, emphasizing the importance of understanding broader ethical and societal contexts. Our approach emphasizes maintaining all potentially relevant variables to allow for more granular fairness assessments and to reduce implicit bias. The findings suggest that the exclusion of sensitive or critical variables may compromise equity between subgroups. In contrast, retaining all relevant variables could reduce implicit bias. Thus, the interdisciplinary approach could provide deeper insight into the ethical implications and compliance with regulatory standards. By integrating a mathematical approach with ethical and social awareness, we suggest more equitable outcomes and responsible AI deployment. This work underscores the necessity of interdisciplinary collaboration in effectively addressing fairness in AI systems aligned with the objectives of the European Union's AI Act, which seeks to promote trustworthy and fair AI systems.

arXiv

Enabling Proactive Spoken Turns via a Generalized Style-Aware Full-Duplex Framework

Tianrui Pan, Qinglin Zhang, Chong Deng, Luyao Cheng, Qian Chen, Wen Wang, Jie Tang, Gangshan Wu, Jie Liu

Compared with half-duplex dialogue systems where the system waits for user turn completion before it responds, natural full-duplex dialogue systems require agents to act proactively in real time, including timely interruptions and backchannels. This creates a key challenge: improving turn timing without sacrificing response quality. To address limitations in realistic proactive turn-taking, we build a generalized style-aware full-duplex framework with three key components. Firstly, we propose LPS-TC, a Lightweight Proactive Speech Turn Controller for plug-and-play integration. It features a fine-grained action space covering both reactive and proactive turn behaviors, enabling half-duplex models with full-duplex capabilities and enhancing existing full-duplex models with superior timing control. Secondly, we construct WildTurn, a large-scale, real-world English dataset containing approximately 2,981 hours of filtered multi-turn stereo conversations from face-to-face and telephone conversations, annotated with five turn-taking and five backchanneling styles. Trained on WildTurn, LPS-TC exhibits rich spoken dynamics that are not captured by existing static full-duplex benchmarks. Thirdly, we introduce a two-tier evaluation scheme that assesses both chunk-level timing precision and turn-level interaction quality under realistic streaming constraints. Our experiments, integrating LPS-TC with half-duplex models like Qwen2.5-Omni and full-duplex models like Freeze-Omni, showcase its superior performance in timing appropriateness and response quality. Our framework also demonstrates fine-grained style controllability and strong generalizability, enabling more natural and human-like spoken interactions.

arXiv

Post-Hoc Trajectory-Risk Certification for Modular LLM-Based Security Agents

Zhenpeng Li

Autonomous security agents operate as staged pipelines, such as classifying network traffic and then attributing attacks to a specific technique. Split conformal prediction gives each stage finite-sample coverage, but deployment requires a trajectory-level guarantee across the full chain. These guarantees do not compose automatically when stages are independently trained and calibrated. Bonferroni allocation is distribution-free but conservative under correlated errors. We show that a natural pairwise-correlation extension to three or more stages is invalid because it gives a lower rather than an upper bound, and derive a valid spanning-tree alternative. We distinguish whether stages are dependent from whether an audit sample is large enough to certify that dependence, and give matching upper and information-theoretic lower sample-complexity bounds. We also show that coarse-to-fine label selection can create near-perfect measured correlation without learned dependence. On a two-stage intrusion-detection pipeline across 6 open LLMs and 2 datasets, removing this artifact reduces measured correlation from near 1 to 0-0.78. A direct audit of trajectory failure becomes 13.7% tighter than Bonferroni once the audit reaches the required sample size, but is worse when undersized. A modular certificate using per-stage certificates and a pairwise overlap bound yields a positive average gain of 0.6%, quantifying the cost of lacking joint access. Same-model, cross-model, and permuted-pairing tests show that residual dependence reflects shared sample difficulty, not shared model representations. Average trajectory coverage across 12 configurations is 92.7% +/- 2.4% at alpha = 0.10. Under cross-dataset deployment, single-step miscoverage reaches 100% even when accuracy remains 78%, showing that distribution shift destroys calibrated confidence before raw accuracy.

arXiv

EA-Graph: Artifact-Anchored Verification Memory for Coding Agents under Upstream Drift

Hwai-Jung Hsu, Cheng-Jan Chi, Hanna Everett

Coding agents increasingly work across sessions, but prose notes can preserve a conclusion without the program state that supported it. After an upstream change, a repository may still build even though earlier verification claims are no longer valid. EA-Graph is an artifact-anchored memory for verification claims. It represents artifacts at sub-path granularity, resolves aliases to leaf definitions, anchors each claim to the content used to establish it, and keeps evidence strength separate from freshness. When replacement content is unavailable, the claim becomes unprovable rather than guessed. EA-Graph is evaluated on generated repositories whose behavior-to-artifact ground truth is known by construction. The task is to classify prior claims as unaffected, affected, or unprovable after value drift, logic drift, and deliberately withheld upstream content. The analysis covers 42 sessions across seven clean worlds, 14 model-world instances, three memory conditions, and two model tiers. In the Haiku round, artifact-anchored memory outscored prose notes and no persistent memory in all seven worlds; each exact paired Wilcoxon comparison yielded p = 0.0156. In the Sonnet round, the anchored condition was perfect, but frequent control ceilings left the preregistered contrasts non-significant. No session fabricated withheld content. These results support a bounded claim: artifact-anchored memory improved the smaller model's provability judgments in this testbed. An exploratory comparison further suggests that structured claim memory may narrow a capability gap by externalizing in-session re-derivation, but it does not establish cross- model equivalence. The study makes no claim about efficiency or repair quality.

arXiv

Compass: Continuously Aligning Social Media Feeds via In-Situ Reflections

Aadit Barua, Leijie Wang, Amy X Zhang

Social media recommendation feeds often optimize for users' immediate impulses rather than preferences they would hold after deeper reflection. Some systems address this misalignment by incorporating users' explicit preferences via a configuration page or in-feed controls instead of just behavioral signals. However, users typically have evolving preferences, and their stated preferences and behavior naturally diverge, necessitating continuous reflection and feed realignment. But existing strategies require the user to take initiative and are often effortful; as a result, in practice they are rarely invoked. We present Compass, a system that aligns a user's feed with their reflective preferences by helping users reflect on and articulate their preferences given their behavior. To enable continuous reflection during everyday browsing, Compass surfaces in-situ reflections via lightweight notifications, while feed alignment is achieved by periodically simulating behavioral signals and directly manipulating feed content. We embedded Compass within YouTube Shorts and compared it against a baseline without continuous support through a 10-day field study (N=15). We found that Compass promoted more reflective and purposeful feed consumption, iterative preference adjustment, and stronger feed alignment, without sacrificing the casual nature of feed browsing.

arXiv

Strategic Evaluation of Planning Strategies for LLM Agents in Cyber-Physical Systems

J. de Curtò, I. de Zarzà

Evaluations of LLM planning agents largely ask whether a task succeeds or a declared plan is followed. In strategic cyber-physical systems, a stronger question is whether the planning architecture remains appropriate after autonomous participants respond and physics constrains the outcome. We introduce a controlled, physics-grounded benchmark built around planning-induced control trajectories: the ordered planning operations and directives through which an execution architecture acts on other agents and the physical process. It implements predefined, sequential, hierarchical, and search executors in a smart-grid demand-response system with 40 heterogeneous prosumers and an independently simulated radial feeder. The LLM is bounded to typed policy declaration and short operator messages, while schedule construction, prosumer dynamics, and power flow remain explicit code. The protocol uses paired forced-mode counterfactuals, common random response draws, and event-level deadline feasibility. Three properties follow. Architecture materially changes outcomes: forced search is the oracle in all five baseline seeds. Execution fidelity needs more than mode agreement: objective substitution holds agreement at 1.0 while increasing voltage shortfall by 2.68x. A 144-scenario, 576-episode bank has feasible oracles from three of the four architectures. A prespecified stress-held-out ridge has mean regret 90.7 (95% interval [73.8, 108.6]) and no detectable value over fixed sequential; applying known deadline feasibility before quality prediction cuts regret to 29.0 and improves over fixed sequential by 61.1. An all-feasible ablation does not beat fixed search, localising the remaining challenge to within-feasible quality selection. A five-model extension separates stress-conditioned, state-blind, and invariant declarers; latency tails show that live feasibility should be treated probabilistically.

arXiv

Interoceptive Attention as Dynamic Homeostatic Prioritization in a Foraging Agent

St John Grimbly, Nicolas Kuske, Evert A. Boonstra, Bruce A. Bassett, Charel van Hoof, Rowan Hodson, Benjamin Rosman, Ryan Smith, Mark Solms, Jonathan P. Shock

Biological systems must regulate competing needs under limited perceptual bandwidth, where sharpening one estimate costs the capacity to sharpen the others. Any fixed-budget system therefore has to decide where to allocate its perceptual precision. We study this in a foraging agent that must keep several bodily needs satisfied to survive, modelled with active inference. At each step it reads its own body-state beliefs, identifies the most-needed channel, and reallocates a fixed budget of interoceptive precision toward it, so that the same precision-shaped likelihood feeds both belief update and planning. In AffectWorld, a four-channel foraging gridworld, this selective allocation more than doubles learning-phase survival at matched budget against a uniform-precision agent ($0.414$ vs $0.199$ across 11 layouts, $n{=}32$ seeds each, paired cluster-bootstrap $p \leq 10^{-4}$). Two further results sharpen the mechanism. The benefit runs through planning as well as perception, since denying the shaped likelihood to the planner alone removes about half of it. It is also need-aligned, since aiming precision at the least-needed channel does worse than spreading it evenly. The attended channel additionally learns its own dynamics about twice as fast, and stays ahead even at matched observation count, a behavioural trace of the same precision routing, visible in learning speed, not survival.

arXiv

A Unified Model for Cross-Domain Clone Detection via Model Merging

Palash R. Roy, Banani Roy, Kevin A. Schneider, Chanchal K. Roy

The growing diversity of code clone types, from syntactic copies to cross-language semantic clones to AI-generated duplicates, has created a fragmentation crisis in clone detection. Current deep learning detectors are domain specialists that degrade significantly outside their training distribution, with F1 drops exceeding 70% across domains. Deploying multiple specialized models is impractical, yet training a single cross-domain detector requires simultaneous access to all training data. To address this, we investigate model merging, a family of post-hoc techniques that operate solely on trained checkpoints. We evaluate parameter merging with five task-vector methods, architecture merging via greedy layer stitching, and cross-tokenizer alignment across four code models, three benchmarks, and twelve configurations. Same-base TIES merging creates effective cross-domain detectors, validated across two model families and three random seeds, reaching 0.865 combined F1 on UniXcoder, 93% of multi-task performance without any training data at the merging step. WUDI achieves the highest in-distribution combined F1 at 0.899, but TIES generalizes better to unseen AI-generated clones, making it our recommended method. Cross-base merging yields only marginal and high-variance gains across all five methods, indicating that task vector compatibility through a shared pre-trained base is the binding factor for effective merging. Merged detectors also outperform zero-shot code LLMs on GPTCloneBench at lower inference cost and generalize up to 4x better than multi-task training to unseen AI-generated clones, suggesting a trade-off between in-domain performance and OOD robustness. This work provides one of the first systematic empirical studies of model merging for software engineering and a practical recipe for building cross-domain clone detectors.

arXiv

MatrAIx: Simulating the World with 8.3 Billion Persona Agents

Xiaomin Li, Yuexing Hao, Jianheng Hou, Jintao Huang, Qianfeng Wen, Shirley Huang, Yifan Liu, Xiaoyi Liu, Yilan Fan, Yijun Wang, Koutian Wu, Ruoqi Gao, Muhammad Ahmed Mohsin, Jing Tang, Brihi Joshi, Heming Liu, Zheyuan Deng, Zonglin Di, Sankalp Jajee, Jiuyao Lu, Zhiwei Zhang, Saksham Kapoor, Ishan Gupta, Yunhan Zhao, Chanwoo Park, Yucheng Lu, Bing Hu, Weihang Xiao, Aravind Mohan, Hanwen Xing, Runyu Zhang, Mihir Kulshreshtha, Yuanda Xu, Qianyu Zhu, Dianzhuo Wang, Yuxin Xiao, Bowen Jiang, Yongye Su, Wenhao Chai, Zuxin Liu, Lawrence Yunliang Chen, Xuandong Zhao, Ethan Ye, Shivam Patel, Jason Xie, Alex Martin Richmond, Weixiang Ding, Emre Okcular, Diya Mathew, Ziheng Wang, Rana M. Shahroz Khan, Zhejian Peng, Fang Wu, Fan Nie, Xinyang Han, Yubin Kim, Jiawei Zhang, Zhenting Qi, Huangyuan Su, Xu Pan, Abinitha Gourabathina, Hyewon Jeong, Hemanth Neelgund Ramesh, Kumail Alhamoud, Kimia Hamidieh, Zidi Xiong, Samuel Schmidgall, Pengrui Han, Yepeng Huang, Yongheng Wang, Bowen Yang, Alex Gu, Yuchu Wang, Akshay Paruchuri, Brenna Li, Hejie Cui, Jiayuan Ding, Chaosheng Dong, Jiahao Wang, Yixuan He, Chi Wang, Pamela Bhattacharya, Tianyi Peng, Paul Pu Liang, Mitchell Gordon, Yilun Du, Marinka Zitnik, James Zou, Prasanna Tambe, Philip Torr, Emily Fox, Asu Ozdaglar, Dawn Song

Human evaluation of AI systems and digital products is costly, slow, and difficult to scale. Offline evaluations are more scalable but often abstract away human diversity and interactive behavior. We therefore introduce MatrAIx, a population-scale simulated-user evaluation infrastructure for testing AI systems and digital products with heterogeneous users. MatrAIx has three core components: First, Persona 8B contains 8.3 billion persona records represented by 1,290 categorical dimensions. Records are either sampled from a dependency graph that preserves correlated attributes or derived from human-authored profiles. We release a quality-filtered coreset of approximately 1 million personas, comprising 599,847 human-grounded and 400,000 synthetic records. Second, the MatrAIx Playground provides four environments in which diverse users evaluate and interact with digital products: Survey, AI Chatbot, Web, and App. Third, MatrAIx provides 1,010 application tasks spanning more than 25 domains, including Commerce, Software, Finance, and Healthcare. We conducted 18,189 evaluation trials across eight representative tasks. Persona agents were powered by three LLMs: Claude Opus 4.8, GPT 5.5, and Claude Haiku 4.5. The resulting feedback captures how decisions and preferences vary across persona backgrounds, including hesitation after a price increase, willingness to continue after an AI assistant fails, and latency tolerance. We conducted two main validation studies: First, a 400-trial controlled study evaluated persona adherence across ten behavioral attributes and all four environments. The declared behavior was expressed or correctly suppressed in 366 trials (91.5%). Second, human and LLM judges evaluated the extraction quality of human-grounded personas. Overall, MatrAIx provides an end-to-end infrastructure for evaluating AI systems and digital products with diverse simulated human users.

arXiv

AgentForge: An Immersive Role-Playing Platform for Learning Agentic Software Engineering

Zihan Fang, Yueke Zhang, Yu Huang

Agentic AI is increasingly used to coordinate planning, implementation, review, and testing in software development, yet it often offers limited transparency into its decisions and interactions. Many such systems also assume that users can effectively guide the AI's decisions and validate its outputs. This assumption poses a particular challenge for novices, who must simultaneously learn how agentic AI works, how to collaborate with it effectively, and how to evaluate its outputs critically. To address this challenge, we present \textit{AgentForge}, an immersive learning system in which novices take on one of four software-engineering roles: Task Planner, Patch Author, Code Reviewer, or Test Runner, within a multi-agent code-repair workflow. In each practice session, the novices perform their chosen role while AI agents perform the remaining three. Through role-based scaffolding and metacognitive support, AgentForge clarifies role-specific responsibilities, makes agent coordination and intermediate artifacts visible, and encourages novices to monitor and evaluate their decisions. In a study with 37 novice developers, participants achieved high task-completion rates with AI-agent support. However, interaction demands differed significantly across practices: the Code Reviewer practice required more interaction turns, reroutes, and completion time ($p_{\mathrm{adj}} = .004$) and was perceived as the most challenging. Participants nevertheless reported significant gains in their understanding of software repair and agent collaboration ($p_{\mathrm{adj}} < .001$). These findings suggest that AgentForge can help novices develop practical software-engineering skills while learning to collaborate with agentic AI more critically and effectively.

arXiv

Monte Carlo Tree Search for Table-to-Multimodal Report Generation

Teng Lin, Zhiyang Zhang, Yuyu Luo, Nan Tang

Automatically generating professional multimodal reports comprising both textual analysis and visual charts from structured tabular data is a critical challenge in data intelligence. Existing methods suffer from fixed linear pipelines and isolated subtask processing, which hinder joint optimization of factual accuracy, visual quality, and narrative coherence. To address these issues, this paper proposes MCTS-Report, a Monte Carlo Tree Search (MCTS)-driven framework that formulates multimodal table-to-report generation as a progressive construction process over a structured search space. The core idea is to decompose report generation into atomic actions, including chapter planning, visualization task identification, chart generation, insight organization, and narrative refinement, each executed by an LLM based on dynamic reasoning conditioned on the current report state. We use an LLM to generate step-by-step reasoning and actions during MCTS, storing the reasoning trajectory in each node for context-aware, coherent report construction. To guide the search, we design a multi-dimensional reward function that jointly evaluates numerical fact consistency (via SQL), chart quality, chart-text alignment, and structural completeness, while incorporating a diversity penalty to suppress repeated charts and a precondition check to prune invalid actions. We also construct MMRBench, a comprehensive benchmark comprising real-world tables from six domains, paired with expert-refined reference report structures and verifiable key insights. Experiments on MMRBench demonstrate that MCTS-Report significantly outperforms strong baselines across structural completeness, numerical accuracy, chart-text alignment, and insight novelty, achieving a 77.9 overall score.

arXiv

The LLM Proposes, the Executive Disposes: A Self-Verifying Agent Instrument that Dissociates Commitment Drift from Binding Drift in Long-Horizon Agents

Mohsen Arjmandi

How do you verify a long-horizon agent when its own state and self-reports are exactly what you cannot trust? We present an agent instrument built so that verification is structural rather than post-hoc. A deterministic Executive owns all belief; a language model may only file typed proposals, and a claim is admitted only when a prediction pre-registered before acting is matched against observation by code. Two properties make the instrument a verifier of its own science, not just of the agent: every run invalidates itself when per-organ write-error, render-size, or salted-canary-echo floors are breached (four of the first eight architecture runs were invalidated, each localizing a real defect); and a render-invisible shadow reference compiles the plan the full system would have committed in every ablation cell, so drift metrics are defined even where the mechanism under test has been removed. Using this instrument we report a clean, single-variable result on a failure every long-horizon agent suffers: ablating the commitment mechanism flips goal-abandonment from 0.00 to 1.00 while binding error stays flat at 0.00 (three seeds per cell, up to 394 reference beats per run, every run gated valid). The binding channel, by contrast, does not reappear as per-beat drift when its repair is ablated -- because binding is code-owned, the failure class is structurally absorbed, its only residue appearing one layer upstream as a collapse in hypothesis formation. We report these under full disclosure that task efficacy is null (zero level completions across 52 gated runs on ARC-AGI-3), pre-registered as a structural defeater. The contribution is a verification methodology for agent development and the drift decomposition it makes measurable.

Keep browsing

More papers from top AI conferences

30+ top conferences, by conference, year, and topic.