Claims: H 2

A 1 · A 2 · A 3 · A 4 · A 5 · A 6 · A 7 · A 8 · B 1 · B 2 · C 1 · C 2 · C 3 · C 4 · C 5 · C 6 · D 1 · D 2 · D 3 · E 1 · E 2 · E 3 · E 4 · F 1 · F 2 · F 3 · G 1 · G 2 · H 1 · H 2 · I 1 · I 2 · I 3 · I 4 · I 5 · J · K · L 1 · L 2 · L 3 · M 1 · M 2 · M 3 · M 4 · N 1 · N 2 · O 1 · O 2 · P 1 · P 2 · P 3 · P 4 · P 5 · Q · R 1 · R 2 · R 3 · S 1 · S 2 · S 3 · S 4 · S 5 · S 6 · S 7 · S 8 · T 1 · T 2 · T 3 · T 4 · T 5 · T 6 · T 7 · T 8 · T 9 · T 10 · U · V · W 1 · W 2 · X · Y · Z · 0-9

• Hispanic and low-income students reported more challenges and technology problems after the shift online than non-Hispanic White and higher-income peers; rural students did not differ - Hispanic and low-income students reported more challenges and technology problems after the shift online than non-Hispanic White and higher-income peers; rural students did not differ
• Hispanic charter students perform similarly to their Hispanic TPS peers in both subjects, while both groups lag the average White TPS student - Hispanic charter students perform similarly to their Hispanic TPS peers in both subjects, while both groups lag the average White TPS student
• Hispanic ELLs proficient before sixth grade performed better in ninth-grade courses than other Hispanic subgroups - Hispanic ELLs proficient before sixth grade performed better in ninth-grade courses than other Hispanic subgroups
• Hispanic parents feel least able to support their children's high school work and report the least school support - Hispanic parents feel least able to support their children's high school work and report the least school support
• Hispanic students graduated at far lower rates than white and Asian ELL students despite similar course performance patterns - Hispanic students graduated at far lower rates than white and Asian ELL students despite similar course performance patterns
• Hispanic students reported higher frequencies of six of the seven non-technology challenges measured in the survey - Hispanic students reported higher frequencies of six of the seven non-technology challenges measured in the survey
• Historical increases in peak performance contradict fixed innate upper limits - Historical increases in peak performance contradict fixed innate upper limits
• Historical legacy of boarding schools and biased attendance responses deepen generational distrust that undermines attendance partnerships with AI/AN families - Historical legacy of boarding schools and biased attendance responses deepen generational distrust that undermines attendance partnerships with AI/AN families
• Historical-method inquiry tasks center on uniformitarianism and taxonomy construction, and evaluating independent lines of inquiry for convergence is entirely absent - Historical-method inquiry tasks center on uniformitarianism and taxonomy construction, and evaluating independent lines of inquiry for convergence is entirely absent
• HME-everyday activities at age 5 showed a protective effect, in contrast to other informal HME models, possibly because items were partially instructional for 5-year-olds - HME-everyday activities at age 5 showed a protective effect, in contrast to other informal HME models, possibly because items were partially instructional for 5-year-olds
• HMeta-d's false positive rate depends on study design and prior specification: independent group priors inflate it, priors on the group difference fix it, and default priors for continuous predictors yield a rate below 5% - HMeta-d's false positive rate depends on study design and prior specification: independent group priors inflate it, priors on the group difference fix it, and default priors for continuous predictors
• HMH Personal Math Trainer use was associated with math gains in grades 4, 5, and 8 but negatively associated in grade 7, and teacher-created assignments showed the strongest association - HMH Personal Math Trainer use was associated with math gains in grades 4, 5, and 8 but negatively associated in grade 7, and teacher-created assignments showed the strongest association
• HOCS clicker questions during learning improve performance on LOCS exam questions more than LOCS clicker questions do - HOCS clicker questions during learning improve performance on LOCS exam questions more than LOCS clicker questions do
• Hofstede's five-dimension cultural model is frequently mapped as a guide to user interface design in HCI - Hofstede's five-dimension cultural model is frequently mapped as a guide to user interface design in HCI
• Holding a strong goal intention alone does not guarantee goal achievement - Holding a strong goal intention alone does not guarantee goal achievement
• Holistic peer feedback correlated significantly with tutor marks while specific-criteria rubric feedback did not in the PEARS system - Holistic peer feedback correlated significantly with tutor marks while specific-criteria rubric feedback did not in the PEARS system
• Holistic retrieval increases situational interest, with better recall performance predicting greater interest, contrary to the knowledge-deprivation hypothesis - Holistic retrieval increases situational interest, with better recall performance predicting greater interest, contrary to the knowledge-deprivation hypothesis
• Holistic retrieval via free-recall improves retention of non-targeted information, but only on the free-recall posttest - Holistic retrieval via free-recall improves retention of non-targeted information, but only on the free-recall posttest
• Holistic retrieval via free-recall raises self-efficacy, fully mediated by performance during retrieval practice - Holistic retrieval via free-recall raises self-efficacy, fully mediated by performance during retrieval practice
• Holt Earth Science inquiry tasks emphasize making multiple observations and developing theories about mechanisms but omit several authentic scientific inquiry features - Holt Earth Science inquiry tasks emphasize making multiple observations and developing theories about mechanisms but omit several authentic scientific inquiry features
• Home literacy environment and child health predict reading readiness at school entry, and socioeconomic status stops being a significant predictor once home literacy environment is entered (suggestive of mediation, not tested) - Home literacy environment and child health predict reading readiness at school entry, and socioeconomic status stops being a significant predictor once home literacy environment is entered (suggestive
• Home literacy environment and family stresses predict attention and behaviour at school entry - Home literacy environment and family stresses predict attention and behaviour at school entry
• Home literacy environment predicts early literacy outcomes, with parent print exposure predicting vocabulary and book-sharing frequency predicting letter knowledge - Home literacy environment predicts early literacy outcomes, with parent print exposure predicting vocabulary and book-sharing frequency predicting letter knowledge
• Home visits are rarely offered and rated lowest of any engagement resource, with Black, Latino, and multilingual parents rating them less favorably - Home visits are rarely offered and rated lowest of any engagement resource, with Black, Latino, and multilingual parents rating them less favorably
• Home visits improved standardized math scores (ES = +0.11) but showed no difference in ELA scores - Home visits improved standardized math scores (ES = +0.11) but showed no difference in ELA scores
• Home visits in funds of knowledge research led teachers to see culture as a dynamic process rather than only dances, food and folklore. - Home visits in funds of knowledge research led teachers to see culture as a dynamic process rather than only dances, food and folklore.
• Home work in arithmetic can be done as well as school work, with overall differences close to zero - Home work in arithmetic can be done as well as school work, with overall differences close to zero
• Homelessness is associated with substantially lower 4-year graduation rates (68% vs. 86% for all students) - Homelessness is associated with substantially lower 4-year graduation rates (68% vs.
• Homework explanations demonstrating understanding predict better isomorphic exam performance across all quartiles - Homework explanations demonstrating understanding predict better isomorphic exam performance across all quartiles
• Homework frequency is more important than homework duration for improving eighth-grade mathematics and science achievement - Homework frequency is more important than homework duration for improving eighth-grade mathematics and science achievement
• Homework management skills partially mediate the relationship between teacher homework involvement and homework behavior - Homework management skills partially mediate the relationship between teacher homework involvement and homework behavior
• Homicides of school-aged children are far less likely to occur on school grounds than elsewhere - Homicides of school-aged children are far less likely to occur on school grounds than elsewhere
• Homogeneous comprehension-needs groups spent less time on engaged reader processes but received better-aligned instruction than heterogeneous groups - Homogeneous comprehension-needs groups spent less time on engaged reader processes but received better-aligned instruction than heterogeneous groups
• Homogeneous CS-major groups reported higher CPSE in collaborative programming than homogeneous non-CS and heterogeneous groups, with no significant difference between the latter two - Homogeneous CS-major groups reported higher CPSE in collaborative programming than homogeneous non-CS and heterogeneous groups, with no significant difference between the latter two
• Homogeneous grouping for reading instruction and heterogeneous grouping for discussion groups each serve distinct purposes - Homogeneous grouping for reading instruction and heterogeneous grouping for discussion groups each serve distinct purposes
• Homogeneous problem perceptions predict better regulation outcomes in collaborative groups
• Hong Kong and Singaporean L2 students underuse discourse connectives overall compared with L1 writers, while Thai students do not differ significantly - Hong Kong and Singaporean L2 students underuse discourse connectives overall compared with L1 writers, while Thai students do not differ significantly
• Hong Kong and Singaporean students are more formal in tone than L1 and Thai students when using adversative and resultive connectives - Hong Kong and Singaporean students are more formal in tone than L1 and Thai students when using adversative and resultive connectives
• Hong Kong fourth graders' OMR self-efficacy and emotions form three distinct motivational-emotional profiles: positive, moderate, and ambivalent - Hong Kong fourth graders' OMR self-efficacy and emotions form three distinct motivational-emotional profiles: positive, moderate, and ambivalent
• Hong Kong students overuse enumerating, resultive and summative discourse connectives relative to L1, Singaporean and Thai writers - Hong Kong students overuse enumerating, resultive and summative discourse connectives relative to L1, Singaporean and Thai writers
• Hong Kong's centralized education reforms provided a favorable macro-context for CSCL classroom innovation despite examination pressure - Hong Kong's centralized education reforms provided a favorable macro-context for CSCL classroom innovation despite examination pressure
• Honours community members perceive a safe and supportive environment that facilitates exploring new ideas - Honours community members perceive a safe and supportive environment that facilitates exploring new ideas
• Honours community members report isolation and negative bias from regular students and lecturers - Honours community members report isolation and negative bias from regular students and lecturers
• Hope is the strongest positive predictor of eudaimonic and overall teacher wellbeing (ρ = .75 and .76) - Hope is the strongest positive predictor of eudaimonic and overall teacher wellbeing (ρ = .75 and .76)
• Hostility and afraid attachment do not meaningfully predict cognitive flexibility, while the other variables do - Hostility and afraid attachment do not meaningfully predict cognitive flexibility, while the other variables do
• Hot EF (CGT risk adjustment) uniquely predicts parent-rated emotional problems, while no cool EF test does - Hot EF (CGT risk adjustment) uniquely predicts parent-rated emotional problems, while no cool EF test does
• Hot executive function (CGT risk adjustment) shows a bell-shaped developmental curve, peaking at age 14 and declining thereafter - Hot executive function (CGT risk adjustment) shows a bell-shaped developmental curve, peaking at age 14 and declining thereafter
• HOTS-oriented teaching improves disposition towards critical thinking subscales including truth-seeking, open-mindedness, self-confidence, and maturity - HOTS-oriented teaching improves disposition towards critical thinking subscales including truth-seeking, open-mindedness, self-confidence, and maturity
• Hours in center-based child care become increasingly positively associated with cognitive scores and negatively associated with behavior scores starting around age 4–4½. - Hours in center-based child care become increasingly positively associated with cognitive scores and negatively associated with behavior scores starting around age 4–4½.
• Household and community factors—parenting knowledge, stress, depression, and resource scarcity—may mediate or constrain SES-related differences in input and learning - Household and community factors—parenting knowledge, stress, depression, and resource scarcity—may mediate or constrain SES-related differences in input and learning
• Household chaos predicts lower prekindergarten math and reading but faster growth in both - Household chaos predicts lower prekindergarten math and reading but faster growth in both
• Household income and maternal work status moderated the childcare–higher education association: formal childcare increased enrollment probability only for low-income and non-working families - Household income and maternal work status moderated the childcare–higher education association: formal childcare increased enrollment probability only for low-income and non-working families
• Households pay substantial recurring sums for private supplementary language education, with Korea at 29.2 trillion won in 2024 and English the leading subject - Households pay substantial recurring sums for private supplementary language education, with Korea at 29.2 trillion won in 2024 and English the leading subject
• Housing tenure, marital status, maternal age and financial difficulties are each significantly associated with ADHD at age 7 in unadjusted analysis - Housing tenure, marital status, maternal age and financial difficulties are each significantly associated with ADHD at age 7 in unadjusted analysis
• Houston charter school ELL students outperformed district-school ELL students in both subjects, while magnet ELL students trailed - Houston charter school ELL students outperformed district-school ELL students in both subjects, while magnet ELL students trailed
• Houston charter school students outperformed Houston other district school students in both reading and math in 2018-19 - Houston charter school students outperformed Houston other district school students in both reading and math in 2018-19
• Houston CMO-affiliated charter students outgrew the state average in both subjects while independent charter students did not differ significantly - Houston CMO-affiliated charter students outgrew the state average in both subjects while independent charter students did not differ significantly
• Houston magnet schools outperformed district schools in reading but not in math in 2018-19 - Houston magnet schools outperformed district schools in reading but not in math in 2018-19
• Houston students made significantly greater math growth than the state average in 2018-19, but reading growth was not significantly different - Houston students made significantly greater math growth than the state average in 2018-19, but reading growth was not significantly different
• How a user engages with AI matters more than whether: structured, deliberate engagement can mitigate or reverse offloading effects - How a user engages with AI matters more than whether: structured, deliberate engagement can mitigate or reverse offloading effects
• How commonly the increased efficiency effect is present in test data is unclear - How commonly the increased efficiency effect is present in test data is unclear
• How People Learn II updates and extends the 2000 How People Learn report by synthesizing research on learning from birth through adulthood in formal and informal settings - How People Learn II updates and extends the 2000 How People Learn report by synthesizing research on learning from birth through adulthood in formal and informal settings
• How school information is presented can affect how parents rank their school choices - How school information is presented can affect how parents rank their school choices
• How students and teachers want to interact with classroom AI largely depends on the teacher-student relationships that exist outside of AI use - How students and teachers want to interact with classroom AI largely depends on the teacher-student relationships that exist outside of AI use
• How students are sorted into classrooms by skill level can affect their achievement as much as the content they are taught - How students are sorted into classrooms by skill level can affect their achievement as much as the content they are taught
• HPT is not a simple shedding of presentist perspectives: multiple selves of one learner intertwine with the target's perspective - HPT is not a simple shedding of presentist perspectives: multiple selves of one learner intertwine with the target's perspective
• HQIM teacher guides differ substantially in the depth of educative guidance they provide for the same pedagogical practice - HQIM teacher guides differ substantially in the depth of educative guidance they provide for the same pedagogical practice
• HRV data show limited predictive validity for academic and performance outcomes in educational contexts - HRV data show limited predictive validity for academic and performance outcomes in educational contexts
• HRV data show moderate concurrent validity as a measure of stress in educational contexts - HRV data show moderate concurrent validity as a measure of stress in educational contexts
• HSE exam takers completed fewer years of school and were more likely to be lunch-eligible and to identify as Black or Hispanic than non-exam takers - HSE exam takers completed fewer years of school and were more likely to be lunch-eligible and to identify as Black or Hispanic than non-exam takers
• HSE learners and workers most wanted more information, skills translation services, job training, mentorship, and money during transitions - HSE learners and workers most wanted more information, skills translation services, job training, mentorship, and money during transitions
• HSE learners and workers saw consolidated records, job matchmaking, convenience, auto-translation, and career search as top LER opportunities - HSE learners and workers saw consolidated records, job matchmaking, convenience, auto-translation, and career search as top LER opportunities
• HSE participants found the LER value proposition unclear, questioning how it differed from Indeed and LinkedIn and feeling excluded by linear-journey prompts - HSE participants found the LER value proposition unclear, questioning how it differed from Indeed and LinkedIn and feeling excluded by linear-journey prompts
• HSGPA has a strong, positive relationship with college graduation that does not vary significantly in its linear slope across high schools - HSGPA has a strong, positive relationship with college graduation that does not vary significantly in its linear slope across high schools
• HSGPA strongly predicts six-year college graduation among four-year college enrollees, with odds of graduating doubling per standard deviation of HSGPA - HSGPA strongly predicts six-year college graduation among four-year college enrollees, with odds of graduating doubling per standard deviation of HSGPA
• HSGPAs have a strong, consistent relationship with college graduation that is larger than school effects, while ACT scores have a weak relationship smaller than high school effects with slopes varying by school - HSGPAs have a strong, consistent relationship with college graduation that is larger than school effects, while ACT scores have a weak relationship smaller than high school effects with slopes varying
• HTKS scores show validity evidence for first graders, associating with teacher-rated social skills and problem behavior - HTKS scores show validity evidence for first graders, associating with teacher-rated social skills and problem behavior
• Huang's meta-analysis of 39 independent samples reported average cross-lagged correlations between self-concept and achievement of roughly .19 to .27 - Huang's meta-analysis of 39 independent samples reported average cross-lagged correlations between self-concept and achievement of roughly .19 to .27
• Human and agent framework revisions concentrate on the mid-level engagement categories (Levels 3–5), indicating greater ambiguity there than at scale endpoints - Human and agent framework revisions concentrate on the mid-level engagement categories (Levels 3–5), indicating greater ambiguity there than at scale endpoints
• Human and AI discourse differ functionally: students lead invitations, connection, and regulation moves, while AI leads reasoning, coordination, and agreement - Human and AI discourse differ functionally: students lead invitations, connection, and regulation moves, while AI leads reasoning, coordination, and agreement
• Human annotators judge later-round CPIG items as improved in complexity but not in difficulty, with modest rater agreement - Human annotators judge later-round CPIG items as improved in complexity but not in difficulty, with modest rater agreement
• Human embodiment in video: perceived social presence benefits learning ratings, but instructor-face inclusion shows no significant learning-performance difference, and learners prefer human over robot presenters with mixed recall - Human embodiment in video: perceived social presence benefits learning ratings, but instructor-face inclusion shows no significant learning-performance difference, and learners prefer human over robot
• Human evaluators rate the framework's simulated solutions higher than the second-best setting across all model settings - Human evaluators rate the framework's simulated solutions higher than the second-best setting across all model settings
• Human evaluators' prioritization of essential features was significantly better for low-quality essays than for average- and high-scoring essays - Human evaluators' prioritization of essential features was significantly better for low-quality essays than for average- and high-scoring essays
• Human expert review is the most credible but slowest and most resource-intensive evaluation method - Human expert review is the most credible but slowest and most resource-intensive evaluation method
• Human experts and LLMs show qualitatively different error patterns in outcome prediction - Human experts and LLMs show qualitatively different error patterns in outcome prediction
• Human Graders of Open-Ended Work Agree With Each Other Less Than Treating One Grade as Ground Truth Implies
• Human inter-rater agreement for FRISCO essay scoring was reliable, with mean kappa 0.78–0.84 and percentage of agreement above 75% in all six dimensions - Human inter-rater agreement for FRISCO essay scoring was reliable, with mean kappa 0.78–0.84 and percentage of agreement above 75% in all six dimensions
• Human inter-rater agreement on the writing rubric is imperfect: no rubric item reaches full agreement and the most discrepant rater generally falls below 80% - Human inter-rater agreement on the writing rubric is imperfect: no rubric item reaches full agreement and the most discrepant rater generally falls below 80%
• Human inter-rater reliability with a single rubric exceeded the ±5-point equivalence margin used for rubric-source comparisons - Human inter-rater reliability with a single rubric exceeded the ±5-point equivalence margin used for rubric-source comparisons
• Human item-level scores on bash command exams are strongly bimodal, concentrated at zero and near-full credit with little intermediate partial credit - Human item-level scores on bash command exams are strongly bimodal, concentrated at zero and near-full credit with little intermediate partial credit
• Human markers detect purchased contract-cheating assignments at low rates, though priming and authorship-analysis software improve detection - Human markers detect purchased contract-cheating assignments at low rates, though priming and authorship-analysis software improve detection
• Human operational raters remain stronger than LLMs at identifying relative strengths (positive feedback) within learner profiles - Human operational raters remain stronger than LLMs at identifying relative strengths (positive feedback) within learner profiles
• Human oversight in co-pilot mode improved Agent Laboratory research output quality ratings from 3.8/10 to 4.38/10 - Human oversight in co-pilot mode improved Agent Laboratory research output quality ratings from 3.8/10 to 4.38/10
• Human oversight remained necessary alongside AI-based tutoring, especially for misconceptions and advanced topics - Human oversight remained necessary alongside AI-based tutoring, especially for misconceptions and advanced topics
• Human performance forms the outer envelope on nearly all MUSE tasks, with the largest deficits in Relative Position, Emotion Detection, and Jigsaw Puzzle - Human performance forms the outer envelope on nearly all MUSE tasks, with the largest deficits in Relative Position, Emotion Detection, and Jigsaw Puzzle
• Human qualitative assessment shows SDXL significantly outperforms SD-v3.5-Medium after fine-tuning, failing on 84 versus 250 of 300 prompts - Human qualitative assessment shows SDXL significantly outperforms SD-v3.5-Medium after fine-tuning, failing on 84 versus 250 of 300 prompts
• Human raters' preferences align with the LLM judge on TutorBench (Pearson r = 0.82 across metric-level win rates) - Human raters' preferences align with the LLM judge on TutorBench (Pearson r = 0.82 across metric-level win rates)
• Human ratings and automated metrics converge on coherence and temporal alignment as the dimensions most enhanced by the dual gatekeeping pipeline - Human ratings and automated metrics converge on coherence and temporal alignment as the dimensions most enhanced by the dual gatekeeping pipeline
• Human relationships are primary means of education, and knowledge of humans' unlearned tendencies is necessary to efficient planning for education in general. - Human relationships are primary means of education, and knowledge of humans' unlearned tendencies is necessary to efficient planning for education in general.
• Human review remains essential after GPT-based codebook generation, producing a final refined SRL codebook of eleven constructs - Human review remains essential after GPT-based codebook generation, producing a final refined SRL codebook of eleven constructs
• Human scoring and feedback for HDR takes about 11 minutes 30 seconds per learner, while GPT-based scoring and feedback is nearly instant - Human scoring and feedback for HDR takes about 11 minutes 30 seconds per learner, while GPT-based scoring and feedback is nearly instant
• Human TAs demonstrated stronger instructor voice, understanding, and helpfulness than AI in responding to student laboratory questions - Human TAs demonstrated stronger instructor voice, understanding, and helpfulness than AI in responding to student laboratory questions
• Human teacher–GenAI collaboration in embodied mathematics learning calls for a complementary interaction model combining AI feedback with human metacognitive guidance - Human teacher–GenAI collaboration in embodied mathematics learning calls for a complementary interaction model combining AI feedback with human metacognitive guidance
• Human tutoring has a medium effect over no tutoring, and in one small study students learned as well when tutors only prompted them as when tutors also explained and gave feedback
• Human-AI collaboration requires refined group-awareness concepts for asymmetrical relationships between users and AI partners - Human-AI collaboration requires refined group-awareness concepts for asymmetrical relationships between users and AI partners
• Human-AI relationship concerns persist across all CAI generations while academic integrity and data privacy are emerging ethical concerns - Human-AI relationship concerns persist across all CAI generations while academic integrity and data privacy are emerging ethical concerns
• Human-AI tutoring increases IXL time on task by 25% with no significant difference between proactive and reactive groups - Human-AI tutoring increases IXL time on task by 25% with no significant difference between proactive and reactive groups
• Human-AI tutoring is associated with a 36% rise in IXL skill proficiency, but the gap widened in favor of reactively tutored students (difference-in-differences preprint) - Human-AI tutoring is associated with a 36% rise in IXL skill proficiency, but the gap widened in favor of reactively tutored students (difference-in-differences preprint)
• Human-GAI paradigms are enacted patterns of engagement rather than fixed individual traits - Human-GAI paradigms are enacted patterns of engagement rather than fixed individual traits
• Human-GenAI pedagogical partnership is positively associated with both cognitive vigilance and cognitive offloading among business students - Human-GenAI pedagogical partnership is positively associated with both cognitive vigilance and cognitive offloading among business students
• Human-human agreement (average κ = 0.644) was lower than the best LLM-LLM agreement (average κ = 0.856) across constructs - Human-human agreement (average κ = 0.644) was lower than the best LLM-LLM agreement (average κ = 0.856) across constructs
• Human-human L2 dialogue has more and shorter turns while AI dialogue has fewer, longer turns with more words per turn - Human-human L2 dialogue has more and shorter turns while AI dialogue has fewer, longer turns with more words per turn
• Human-LLM agreement is moderated by code properties, multi-model consensus, and model-reported confidence - Human-LLM agreement is moderated by code properties, multi-model consensus, and model-reported confidence
• Human-LLM agreement on a complex multi-label codebook falls well below human-human agreement, while LLM-LLM agreement is comparable to human-human agreement - Human-LLM agreement on a complex multi-label codebook falls well below human-human agreement, while LLM-LLM agreement is comparable to human-human agreement
• Human-LLM agreement remains low across constructs (κ = 0.000 to 0.419), with self-efficacy showing the best alignment and Prior KSAs none - Human-LLM agreement remains low across constructs (κ = 0.000 to 0.419), with self-efficacy showing the best alignment and Prior KSAs none
• Human-resource roadblocks rose sharply from 2009 to 2017: difficulty removing poor teachers increased from 39 to 48 percent, and concern over recruiting and hiring the right teachers nearly quadrupled from 12 to 44 percent - Human-resource roadblocks rose sharply from 2009 to 2017: difficulty removing poor teachers increased from 39 to 48 percent, and concern over recruiting and hiring the right teachers nearly quadrupled
• Human-shadow prompting generates broad psychometric variance while AI-assistant responses are uniform - Human-shadow prompting generates broad psychometric variance while AI-assistant responses are uniform
• Human-written AGENTS.md context files improve agent performance by approximately 4 per cent, while LLM-generated context files decrease performance by 3 per cent and increase token cost by over 20 per cent, as the article reports citing recent research - Human-written AGENTS.md context files improve agent performance by approximately 4 per cent, while LLM-generated context files decrease performance by 3 per cent and increase token cost by over 20 per
• Humanities students report higher perceived usefulness, ease of use, and behavioral intention for AI-assisted English tools than STEM students, with small effect sizes - Humanities students report higher perceived usefulness, ease of use, and behavioral intention for AI-assisted English tools than STEM students, with small effect sizes
• Humanization works by systematically suppressing detector-salient surface cues — long-token ratio, AWL density, and lexical diversity — rather than fragmenting syntax - Humanization works by systematically suppressing detector-salient surface cues — long-token ratio, AWL density, and lexical diversity — rather than fragmenting syntax
• Humanizer legitimacy performances show leakage: unendorsed logos, inactive social media, and no registered business address - Humanizer legitimacy performances show leakage: unendorsed logos, inactive social media, and no registered business address
• Humanizer sites deploy mystification through opaque algorithm references, scale claims, and unexplained success-rate claims - Humanizer sites deploy mystification through opaque algorithm references, scale claims, and unexplained success-rate claims
• Humanizer sites frame AI humanization as a rational, ethically defensible response to flawed surveillance and false positives - Humanizer sites frame AI humanization as a rational, ethically defensible response to flawed surveillance and false positives
• Humanlike AI co-writing (real-time suggestions and an agentic cursor) evoked feelings of surveillance, judgment and eeriness, and the moving cursor distracted writers - Humanlike AI co-writing (real-time suggestions and an agentic cursor) evoked feelings of surveillance, judgment and eeriness, and the moving cursor distracted writers
• Humans are more certain when judging that facilitation is not needed than when judging it is - Humans are more certain when judging that facilitation is not needed than when judging it is
• Human–GenAI collaborative interactions show promise for supporting metacognition and sociometacognition, based on an exploratory expert case study - Human–GenAI collaborative interactions show promise for supporting metacognition and sociometacognition, based on an exploratory expert case study
• Human–human pair programming fostered the highest perceptions of collaboration and social presence, surpassing both AI-assisted and individual conditions - Human–human pair programming fostered the highest perceptions of collaboration and social presence, surpassing both AI-assisted and individual conditions
• Hungarian transformer models achieve slightly lower overall score (about 68%) but better minority-class sensitivity than shallow models - Hungarian transformer models achieve slightly lower overall score (about 68%) but better minority-class sensitivity than shallow models
• Hunter-gatherer communities also view schooling positively, as a means of asserting rights, socioeconomic empowerment, and new social relationships - Hunter-gatherer communities also view schooling positively, as a means of asserting rights, socioeconomic empowerment, and new social relationships
• Hurst exponents are unrelated to the frequency of interacting with any specific game-based feature - Hurst exponents are unrelated to the frequency of interacting with any specific game-based feature
• Hurst exponents do not significantly predict retention self-explanation quality over and above pretest scores - Hurst exponents do not significantly predict retention self-explanation quality over and above pretest scores
• Hurst exponents predict daily training self-explanation quality over and above pretest self-explanation ability - Hurst exponents predict daily training self-explanation quality over and above pretest self-explanation ability
• Hybrid approaches combining LLMs with traditional knowledge-tracing algorithms outperform either method alone for learner modelling - Hybrid approaches combining LLMs with traditional knowledge-tracing algorithms outperform either method alone for learner modelling
• Hybrid approaches with smaller masked groups attending part-time dramatically reduce total likely infections, and most outside infections produce zero additional in-school infections - Hybrid approaches with smaller masked groups attending part-time dramatically reduce total likely infections, and most outside infections produce zero additional in-school infections
• Hybrid feedback outperforms direct GenAI feedback on conceptual learning; the reflective–direct contrast is positive but non-significant - Hybrid feedback outperforms direct GenAI feedback on conceptual learning; the reflective–direct contrast is positive but non-significant
• Hybrid quantum optimization is applicable to school aggregation as a proof of concept, with classical optimization currently advantaged - Hybrid quantum optimization is applicable to school aggregation as a proof of concept, with classical optimization currently advantaged
• Hybrid training fosters far transfer of metacognitive skills to a new cognitive strategy, though one hybrid condition did not beat control - Hybrid training fosters far transfer of metacognitive skills to a new cognitive strategy, though one hybrid condition did not beat control
• Hybrid training improves fifth-graders' application of metacognitive skills in near-transfer tasks compared to non-hybrid and control training - Hybrid training improves fifth-graders' application of metacognitive skills in near-transfer tasks compared to non-hybrid and control training
• Hybrid work must be reconceptualised to include the human/AI duality alongside the presence/virtual continuum - Hybrid work must be reconceptualised to include the human/AI duality alongside the presence/virtual continuum
• Hybridization of authorship blends human and AI-generated content, often without learners' explicit awareness of the boundary, normalizing formulaic rhetorical structures - Hybridization of authorship blends human and AI-generated content, often without learners' explicit awareness of the boundary, normalizing formulaic rhetorical structures
• Hyper-reference dictionary users consulted vocabulary twice as often as paper-dictionary users, though comprehension did not differ significantly - Hyper-reference dictionary users consulted vocabulary twice as often as paper-dictionary users, though comprehension did not differ significantly
• Hyperactivity and conduct problems were not statistically significant predictors of reading comprehension in this sample - Hyperactivity and conduct problems were not statistically significant predictors of reading comprehension in this sample
• Hypermedia learners show the lowest total engagement in learning processes, devoting most processing to superficial activities - Hypermedia learners show the lowest total engagement in learning processes, devoting most processing to superficial activities
• Hyperparameter sensitivity in PKT is an inherent characteristic of programming interaction data, varying across models and across assignments within the same model - Hyperparameter sensitivity in PKT is an inherent characteristic of programming interaction data, varying across models and across assignments within the same model
• Hyperparameter sweeps reproduce the lesson's claims: γ = 0 never solves the task while ε = 1 still converges - Hyperparameter sweeps reproduce the lesson's claims: γ = 0 never solves the task while ε = 1 still converges
• Hypertension is associated with poorer Stroop Word and Color performance in healthy older adults - Hypertension is associated with poorer Stroop Word and Color performance in healthy older adults
• Hypertextuality extends intertextuality into the digital world, foreshadowing connectivism - Hypertextuality extends intertextuality into the digital world, foreshadowing connectivism
• Hypotheses sharing relates positively to diagnostic justification and diagnostic efficiency - Hypotheses sharing relates positively to diagnostic justification and diagnostic efficiency
• Hypotheses that specific robot appearances would be trusted more in matching tasks (humanoids for teaching, task-oriented robots for instructions, soft robots for personal disclosure) were not confirmed - Hypotheses that specific robot appearances would be trusted more in matching tasks (humanoids for teaching, task-oriented robots for instructions, soft robots for personal disclosure) were not confirm
• Hypothesis specificity differed by task type: intermediate participants stated fewer partially-specified hypotheses than abstract participants, but more fully-specified hypotheses than concrete participants - Hypothesis specificity differed by task type: intermediate participants stated fewer partially-specified hypotheses than abstract participants, but more fully-specified hypotheses than concrete partic
• Hypothesis-based predictors match or beat black-box difficulty regressors on unseen questions - Hypothesis-based predictors match or beat black-box difficulty regressors on unseen questions
• Hypothesized ATI effects of individualized explanations based on typical mistakes were not supported; exploratory benefits emerged for high-resource vocational-school students - Hypothesized ATI effects of individualized explanations based on typical mistakes were not supported; exploratory benefits emerged for high-resource vocational-school students
• Hypothesized indirect effects from technology acceptance to engagement through reflective use, PSE, and the serial RU–PSE path are not significant (null result) - Hypothesized indirect effects from technology acceptance to engagement through reflective use, PSE, and the serial RU–PSE path are not significant (null result)