{"total":13,"items":[{"citing_arxiv_id":"2606.30549","ref_index":27,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"To Tab or Not to Tab: Measuring Critical Engagement in AI Code Completion Tools Using Behavioral Signals and Attention Checks","primary_cat":"cs.HC","submitted_at":"2026-06-29T16:47:46+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Clover tool and behavioral taxonomy show tab-accept rates linked to lower attention-check scores and dwell time linked to higher scores in AI-assisted programming tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.02933","ref_index":71,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Characterization and Effects of CS2 Learning with GenAI, Visualization, and Human Support","primary_cat":"cs.HC","submitted_at":"2026-06-01T22:31:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"GenAI produced larger self-efficacy gains but noticeably lower learning outcomes than live tutoring, with visualizations underused and GenAI facing barriers on advanced topics.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.14849","ref_index":38,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Beliefs and Misconceptions around Integrated Conversational AI","primary_cat":"cs.HC","submitted_at":"2026-05-14T13:57:31+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Qualitative study of 20 users of integrated browser conversational AI found that citations raise trustworthiness without verification and that users apply existing LLM and search perceptions to prompting strategies.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.10702","ref_index":39,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"ChatGPT: Friend or Foe When Comprehending and Changing Unfamiliar Code","primary_cat":"cs.SE","submitted_at":"2026-05-11T15:17:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Developers using AI showed the same core problem-solving behaviors as those without but differed in how they became stuck and recovered, with AI helping or hindering in specific cases.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"It has already been noted that GenAI tools can help in this process, for instance, by explaining existing code [11], providing relevant information on demand (e.g., details of an external API) [32], and suggesting code (e.g., [30]). How the use of GenAI potentially alters the problem-solving behaviors developers exhibit in addressing a change task, however, is little studied [39]. We report on the results of an exploratory laboratory study with ten participants in which we compared the problem-solving behaviors employed by participants who had access to a GenAI tool (ChatGPT) to those who did not. We specifically sought to answer the following two research questions: RQ1 How does the use of AI tools impact how developers prob-"},{"citing_arxiv_id":"2605.05767","ref_index":74,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Priming, Path-dependence, and Plasticity: Understanding the molding of user-LLM interaction and its implications from (many) chat logs in the wild","primary_cat":"cs.HC","submitted_at":"2026-05-07T07:00:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Large-scale analysis of wild LLM chat logs finds that user interaction patterns stabilize quickly after initial use and correlate with long-term outcomes like retention, creating an agency paradox of limited exploration in unconstrained systems.","context_count":1,"top_context_role":"background","top_context_polarity":"unclear","context_text":"InProceedings of the 2025 CHI Conference on Human Factors in Computing Systems (CHI '25). Association for Computing Machinery, New York, NY, USA, Article 481, 17 pages. doi:10.1145/3706598.3713120 [73] Barry Schwartz. 2015. The paradox of choice.Positive psychology in practice: Promoting human flourishing in work, health, education, and everyday life(2015), 121-138. [74] John R Searle. 1969.Speech acts: An essay in the philosophy of language. Cam- bridge university press. [75] Orit Shaer, Angelora Cooper, Osnat Mokryn, Andrew L Kun, and Hagit Ben Shoshan. 2024. AI-Augmented Brainwriting: Investigating the use of LLMs in group ideation. InProceedings of the 2024 CHI Conference on Hu- man Factors in Computing Systems(Honolulu, HI, USA)(CHI '24)."},{"citing_arxiv_id":"2604.18948","ref_index":66,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Relationships Between Trust, Compliance, and Performance for Novice Programmers Using AI Code Generation","primary_cat":"cs.HC","submitted_at":"2026-04-21T00:52:29+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Among novice programmers using AI code generators, trust did not predict compliance with suggestions, while performance correlated with both compliance and increased subsequent trust.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.18883","ref_index":44,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Choose Your Own Adventure: Non-Linear AI-Assisted Programming with EvoGraph","primary_cat":"cs.HC","submitted_at":"2026-04-20T22:05:09+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"EvoGraph turns linear AI-assisted programming into a manipulable graph of branching histories, reducing cognitive load and enabling better iteration according to a user study with 20 developers.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Becker, Bailey Kimmel, Jared Wright, and Ben Briggs. 2024. The Widening Gap: The Benefits and Harms of Generative AI for Novice Programmers. InProceedings of the 2024 ACM Conference on International Computing Education Research - Volume 1(Melbourne, VIC, Australia)(ICER '24). Association for Com- puting Machinery, New York, NY, USA, 469-486. doi:10.1145/3632620.3671116 [44] Kevin Pu, Daniel Lazaro, Ian Arawjo, Haijun Xia, Ziang Xiao, Tovi Grossman, and Yan Chen. 2025. Assistance or Disruption? Exploring and Evaluating the Design and Trade-offs of Proactive AI Programming Support. InProceedings of the 2025 CHI Conference on Human Factors in Computing Systems (CHI '25). Association for Computing Machinery, New York, NY, USA, Article 152, 21 pages."},{"citing_arxiv_id":"2604.18538","ref_index":61,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Fast and Forgettable: A Controlled Study of Novices' Performance, Learning, Workload, and Emotion in AI-Assisted and Human Pair Programming Paradigms","primary_cat":"cs.HC","submitted_at":"2026-04-20T17:33:13+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Novices performed better and reported lower workload with GitHub Copilot than with human partners, but human partners produced more positive emotions and a smaller drop in retest performance after one week.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Pairs of weak partners struggled to make progress together; AI helped with this, but also dominated the team in the process. Participants felt that they could communicate ineffectively and still obtain helpful results. This marks a potential harm to the metacognitive skill of analyzing and expressing a question or goal, which has previously been considered as helped by the act of prompting [ 61]. Finally, Fan et al. reported a quasi-experimental between-subjects study of 234 students in a web development course who programmed in either solo, human-human, or human-AI paradigms [25]. They found that performance was comparably higher in both paired conditions compared to the solo condition. Programming anxiety was most reduced by the human-AI paradigm, but"},{"citing_arxiv_id":"2604.18530","ref_index":61,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"OGER: A Robust Offline-Guided Exploration Reward for Hybrid Reinforcement Learning","primary_cat":"cs.AI","submitted_at":"2026-04-20T17:26:00+00:00","verdict":"CONDITIONAL","verdict_confidence":"UNKNOWN","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Novice programmers completed more tasks with lower workload using GitHub Copilot versus a human partner, but reported significantly more positive and arousing emotions with the human teammate.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.16393","ref_index":52,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"How Do Developers Interact with AI? An Exploratory Study on Modeling Developer Programming Behavior","primary_cat":"cs.SE","submitted_at":"2026-03-28T14:37:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Developers using AI assistants exhibit more stable emotions and greater focus on code creation, evaluation, and verification, captured in a new four-dimensional S-IASE model from retrospective labeling of screen recordings, surveys, and interviews.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.16365","ref_index":10,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"\"CS 1.5\": An Experience Report on Integrating CS1 and Discrete Structures for the AI Era","primary_cat":"cs.CY","submitted_at":"2026-03-21T21:15:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"An experience report describes the design and delivery of an integrated CS1-plus-Discrete-Structures studio course called CS 1.5 that treats AI as a partner and emphasizes theoretical foundations through code comprehension projects.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2602.21697","ref_index":48,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"EditFlow: Benchmarking and Optimizing Code Edit Recommendation Systems via Reconstruction of Developer Flows","primary_cat":"cs.SE","submitted_at":"2026-02-25T09:02:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"EditFlow reconstructs temporal developer editing flows from code changes to benchmark and optimize AI code edit recommenders so they align with natural incremental reasoning rather than static snapshots.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2410.01026","ref_index":73,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Understanding the Human-LLM Dynamic: A Literature Survey of LLM Use in Programming Tasks","primary_cat":"cs.SE","submitted_at":"2024-10-01T19:34:46+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"A survey of user studies on LLM use in programming that identifies interaction behaviors, mixed benefits and weaknesses, and factors influencing human and task performance.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}