Pith. sign in

REVIEW 3 major objections 7 minor 1 cited by

EAVIT: Efficient and Accurate Human Value Identification from Text data via LLMs

T0 review · 3 major / 7 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Human value identification from text becomes both cheaper and more accurate when a local model pre-selects candidate values before the LLM arbitrates the remaining candidates.

desk verdict Plausible, useful two-stage value-identification pipeline with strong reported F1 and token savings, but test-set threshold selection means the headline margins are not yet proven. read the letter →

arxiv 2505.12792 v1 pith:DQQERVHJ submitted 2025-05-19 cs.CL

Wenhao Zhu , Yuhang Xie , Guojie Song , Xin Zhang This is my paper · ORCID
classification cs.CL
keywords humanvalueidentificationlargelanguagemodelsmulti-labelclassificationSchwartztheoryofbasicvaluesdetectorcandidatesetgenerationexplanation-basedfine-tuningLLMpromptingefficiency
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper sets out to show that identifying human values in text, a multi-label classification against a long, definition-heavy taxonomy, is best done by splitting the work between a small local model and a large online one. The local model proposes a short list of candidate values, and the LLM decides only within that list, so the prompt stays short. On the Touché23-ValueEval test set the full EAVIT pipeline reaches F1 0.69, compared with 0.56 for the best SemEval-2023 system and 0.58 for GPT-4o with sequential chain-of-thought prompting, while using 0.45k tokens per sample instead of 2.4k to 3.6k. The paper argues that both the accuracy gain and the cost reduction follow from replacing the full value-system definition with concise candidate-set prompts, and it attributes the detector's usefulness to explanation-based fine-tuning, definition reflection, and LLM-generated training data. A virtual-individual experiment suggests the same approach can recover a person's values from ordinary text more reliably than direct LLM prompting.

What carries the argument

The mechanism that carries the argument is the candidate-value-set gate. A local value detector, Llama2-13b-chat fine-tuned with QLoRA and explanation-based supervision, is sampled $L=5$ times per input, producing for each value $V_i$ a mean sampled vote. The pipeline then splits the value set into three zones: values whose vote share is above $p_{high}$ are accepted, values whose vote share is below $p_{low}$ are discarded, and the middle band between $p_{low}$ and $p_{high}$ enters the candidate set $S(T)$ that the LLM arbitrates. The prompt sent to the LLM contains only the definitions of the values in $S(T)$, which averages 3.3 per example, so the context shrinks from the full 2.5k-token value system to roughly 0.45k tokens. This gate is where both savings arise: it removes the long-context degradation the paper attributes to longer prompts, and it cuts API cost by a large factor. Everything else, including data augmentation, definition reflection, and explanation-based fine-tuning, exists to make the gate's probabilities reliable enough that the cheap local model can be trusted outside the band.

What would settle it

Run the fixed EAVIT thresholds $p_{low}=0.2$ and $p_{high}=0.8$ on a fresh, out-of-distribution corpus with human-annotated Schwartz values and compare candidate-set recall and final F1 against direct GPT-4o sequential CoT prompting; if true positive values below $p_{low}$ are common enough to leave EAVIT behind the direct baseline, the accuracy half of the central claim is refuted.

Watch

Extended reading notes

Core claim

The paper's central claim is that a short prompt built from a candidate set can beat a long prompt built from the full value system for this task, and that the local detector is what makes the short prompt sufficient. EAVIT samples the value detector five times per input, computes the mean sampled vote for each value $V_i$, and uses thresholds $p_{low}=0.2$ and $p_{high}=0.8$; values above the upper threshold are accepted, values below the lower threshold are discarded, and the rest form the candidate set $S(T)$ that GPT-4o or GPT-4o-mini arbitrates using only the definitions of the values in $S$. On the public benchmarks this yields the reported F1 gains, $0.69 \pm 0.02$ on the Touché23-ValueEval test set versus 0.56 for the SemEval-2023 best system and 0.58 for GPT-4o sequential CoT prompting, while the number of input tokens per sample drops from 2.4k to 3.6k down to 0.45k, about one-fifth to one-eighth of the original. The paper also claims that the candidate gate is stable: the final LLM step suppresses the detector's sampling variance, and the accuracy gains persist on Webis-ArgValues-22 and the ValueNet-augmented dataset.

Load-bearing premise

The pipeline works only if the local detector's five-sample vote can be trusted to say that values scoring under $p_{low}=0.2$ are irrelevant and values scoring over $p_{high}=0.8$ are relevant, without the LLM ever checking them, and those threshold values were themselves chosen to maximize F1 on the test set.

Editorial extensions

If this is right

  • Value tagging at web scale becomes financially feasible because the API token budget is roughly one-fifth to one-eighth of direct prompting, so millions of argumentative texts can be screened for Schwartz values at a fraction of the previous cost.
  • The method is not tied to the Schwartz taxonomy: any completely specified value system with textual definitions and labeled examples can drive the same detector-plus-arbiter split, per the paper's stated generalization.
  • For LLM alignment, cheap and accurate value identification makes it practical to audit model outputs for which human values they invoke or oppose, rather than relying on expensive full-context annotations.
  • For psychology, the virtual-individual case study indicates that passively generated texts can stand in for active questionnaires, with EAVIT predicting the questionnaire-derived values more accurately than the LLM baselines it is compared with.
  • Because the final LLM step collapses the detector's output variance, the pipeline is suitable for repeated large-scale measurement where output stability matters.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The reported thresholds were selected by maximizing F1 on the test set, so the headline numbers are likely optimistic; a stronger evaluation would tune $p_{low}$ and $p_{high}$ on a separate development split and report the untouched-test numbers.
  • A testable extension is to replace fixed thresholds with a calibrated uncertainty measure such as vote entropy or margin over the five samples, which could make the gate adaptive to new corpora without re-tuning.
  • The same prefilter-plus-arbiter pattern should transfer to other definition-heavy multi-label tasks, including moral foundations tagging, legal code classification, and scientific claim typing, wherever the label definitions are long relative to the text.
  • The passive-value-measurement idea is demonstrated only with GPT-simulated individuals; the decisive experiment would pair real people's social media or interview text with their questionnaire answers on the same individuals.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 7 minor

Summary. The paper proposes EAVIT, a two-stage pipeline for multi-label human value identification. A local Llama2-13b-chat model, fine-tuned with QLoRA using explanation-based prompts and augmented with GPT-4o-mini generated data, serves as a value detector: it is sampled L=5 times per input, producing a per-value relevance probability. Thresholds plow and phigh split the value set into accepted positives (p>phigh), accepted negatives (p<plow), and a candidate set (plow≤p≤phigh) whose definitions are sent to an online LLM (GPT-4o or GPT-4o-mini) for final decisions. Experiments on Touché23-ValueEval and Webis-ArgValues-22 report state-of-the-art F1 (up to 0.69±0.02 on the Touché test set versus 0.56 for the SemEval-2023 best system and 0.58 for GPT-4o sequential CoT prompting) with an LLM token cost of 0.45k per sample versus 2.4k–3.6k for direct prompting. The paper also reports results on ValueNet and a case study that compares value identification from simulated social-media text with psychological questionnaire scores. The authors attribute the gains to the detector's filtering effect, explanation-based fine-tuning, and targeted data generation.

Significance. The core idea of using a cheap local model to filter candidate values and to shorten the definition context for a black-box LLM is simple, practical, and likely to be useful beyond this specific task. The paper includes several strengths: EAVIT's own results are averaged over three runs with standard deviations; the ablation study (Figure 6) isolates the contribution of explanation-based fine-tuning and data generation; the token-cost evaluation (Table 4) is a useful addition; and the virtual-individual case study, while exploratory, connects the method to a psychological validation setting. If the results survive a proper evaluation, EAVIT would provide a cost-effective alternative to direct LLM prompting for value identification, with a five- to eight-fold token reduction and a simultaneous accuracy improvement. However, the headline claim is currently weakened by the fact that the banding thresholds were selected on the test set, which biases both the F1 margin and the token savings; and the baselines lack error bars, so the comparisons are not fully quantified.

major comments (3)
  1. [Appendix 8.2, Table 5; Section 4.2] The thresholds plow=0.2 and phigh=0.8 are selected by comparing four threshold pairs on the Touché23-ValueEval test set and taking the pair with the highest F1. Since the banding rule determines both the accuracy gain (Table 1: 0.66–0.69 vs 0.56–0.58) and the token reduction (0.45k vs 2.4–3.6k), the headline results are selected optima rather than unbiased estimates for new data. Please re-select thresholds on a validation split or with cross-validation, and report the test-set performance with thresholds fixed before any test-set access, including the variability of the selected thresholds across runs.
  2. [Tables 1 and 2; Section 5.1] Only the EAVIT rows in Tables 1 and 2 carry standard deviations (over three runs); all baseline numbers are point estimates. The claim that EAVIT 'consistently outperforms' prior methods is thus not fully quantified, and at least one comparison is a tie (ValueNet test accuracy 0.78 for both EAVIT and GPT-4o-mini sequential prompting, Table 2). Please report error bars for all methods, or a paired significance test for the main comparisons (e.g., bootstrap over test items or McNemar's test) so the reader can judge whether the margins are statistically meaningful.
  3. [Section 4.2 and Appendix 8.2] The threshold ablation in Table 5 considers only four threshold pairs on a single test set, and no sensitivity analysis is provided for the other datasets or for the claim that the method transfers to 'any completely defined value system' (Section 8.3). Please report the same threshold grid (or a robustness curve around the chosen pair) on Webis-ArgValues-22 and ValueNet, and discuss whether the optimal thresholds are dataset-dependent, since the validity of the banding rule is the main load-bearing assumption of the pipeline.
minor comments (7)
  1. [Abstract and Section 1] 'reduces the number of input tokens by up to 1/6' is ambiguous and inconsistent with Table 1, where the reductions are to about 1/5–1/8 of the direct-prompting token count. Please rephrase to 'down to roughly 1/5–1/8' or give the exact factors.
  2. [Section 5.2] 'strong non-falsifiability' is not a sensible advantage; the authors likely mean unobtrusiveness or low social-desirability bias. Please reword.
  3. [Section 4.1 and throughout] There are several typographical errors, including 'GPT-4o-miniGPT-4o-mini' and garbled Unicode fragments (e.g., in the discussion of Figure 3); the manuscript needs a full proofreading pass.
  4. [Table 2] Table 2 reports only Accuracy on ValueNet, while Table 1 uses F1 (the official metric for this multi-label task); please report F1 or another multi-label metric on ValueNet as well.
  5. [Section 5.1] The sentence 'We report the average and std of 3 random individual runs' follows the EAVIT setup; please specify explicitly whether this statement applies to all rows of Table 1 or only to EAVIT, and state the number of runs for the baselines.
  6. [Section 4.3] The average candidate set size (3.3 for Touché23-ValueEval) should be accompanied by its distribution, since the token cost (and thus the efficiency claim) depends directly on this quantity.
  7. [Section 3.2] The citation for GPT-2 points to Brown et al. (the GPT-3 paper); please cite the GPT-2 paper (Radford et al.) or clarify the intended model.

Circularity Check

1 steps flagged · score 5.0 of 10

Threshold selection on the test set (Table 5) makes the headline F1 a selected optimum; otherwise the derivation is self-contained.

  1. fitted input called prediction [Appendix 8.2 (Table 5) and Section 5.1 (Table 1): threshold selection p_low=0.2, p_high=0.8]
    "Table 5 is an ablation study of the thresholds plow and phigh in EA VIT, which control the selection of candidate values to LLM for final identification 4.2. According to the results, the current experimental set plow = 0.2 and phigh = 0.8 is the optimal setting."

    The main EAVIT results in Table 1 (F1 0.69±0.02 on the Touché23-ValueEval test set with GPT-4o, and 0.66 with GPT-4o-mini, with 0.45k tokens) are produced using p_low=0.2, p_high=0.8. Appendix Table 5 shows that these thresholds were selected by evaluating four threshold pairs on the same ValueEval test set and taking the pair with the highest F1 (0.67 vs 0.59, 0.65, 0.62). Thus the reported test F1 is the maximum of a small search over thresholds fitted to the same test labels, not an unbiased prediction of a pre-specified method's performance. The token-saving claim is also tied to this test-selected operating point, since the band determines how many values enter the LLM prompt.

full rationale

The derivation is otherwise self-contained: the value detector is fine-tuned on human-annotated datasets plus generated explanations, and the final LLM arbitration is evaluated against human labels. No load-bearing self-citation or imported uniqueness theorem was found; [Ren et al., 2024] is cited only as an example of another value system. The one concrete reduction is the threshold choice: p_low and p_high are selected by maximizing F1 on the same test set used for the reported headline performance (Table 5), so part of the reported margin is a selected optimum rather than an unbiased prediction. This affects both the F1 gain and the token-cost claim, because the selected band directly sets the candidate-set size and hence the LLM context length. The central method is not circular by definition, but the test-set selection step should be corrected with validation-based threshold selection before the SOTA margin is taken at face value.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The method introduces no new physical or conceptual entities. It relies on two fitted thresholds and a hand-chosen sampling count, plus assumptions about the value taxonomy and the reliability of LLM-generated training data.

free parameters (3)
  • p_low = 0.2
    Lower threshold for candidate set membership, chosen as optimal on the ValueEval test set (Appendix 8.2, Table 5). Affects how many values go to the LLM and hence accuracy and token cost.
  • p_high = 0.8
    Upper threshold; values above this are accepted by the local detector without LLM review. Also selected on the test set.
  • L (sampling rounds) = 5
    Number of stochastic detector outputs per input used to estimate relevance probabilities; chosen by hand as a balance of stability and cost (Section 4.2).
assumptions (4)
  • domain assumption Schwartz value taxonomy is the correct ground-truth label system
    The paper follows Kiesel et al. and uses the 20 level-2 Schwartz values as the target labels; if this taxonomy is not appropriate for the text domain, the task itself is questionable.
  • domain assumption The value detector's sampled output frequencies approximate true relevance probabilities
    Section 4.2 states 'Our basic assumption is that the value detector will produce outputs that have some random deviations compared to the correct results.' The candidate set method depends on this.
  • domain assumption GPT-4o-mini-generated training data and explanations are reliable enough to improve the detector
    Section 4.1 uses LLM-generated explanations and synthetic instances for training; if the generator introduces systematic label noise, the detector's probabilities are biased.
  • domain assumption Long contexts degrade LLM performance as claimed
    Motivates token reduction; based on Liu et al. 2023, not re-derived here.

how reviews work

0 comments
Cite this review

Pith. "Pith review of EAVIT: Efficient and Accurate Human Value Identification from Text data via LLMs." pith.science (2026). https://pith.science/paper/DQQERVHJ

@misc{pith2026250512792,
  author       = {Pith},
  title        = {Pith review of: EAVIT: Efficient and Accurate Human Value Identification from Text data via LLMs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DQQERVHJ}},
  note         = {Machine review of arXiv:2505.12792}
}
read the original abstract

The rapid evolution of large language models (LLMs) has revolutionized various fields, including the identification and discovery of human values within text data. While traditional NLP models, such as BERT, have been employed for this task, their ability to represent textual data is significantly outperformed by emerging LLMs like GPTs. However, the performance of online LLMs often degrades when handling long contexts required for value identification, which also incurs substantial computational costs. To address these challenges, we propose EAVIT, an efficient and accurate framework for human value identification that combines the strengths of both locally fine-tunable and online black-box LLMs. Our framework employs a value detector - a small, local language model - to generate initial value estimations. These estimations are then used to construct concise input prompts for online LLMs, enabling accurate final value identification. To train the value detector, we introduce explanation-based training and data generation techniques specifically tailored for value identification, alongside sampling strategies to optimize the brevity of LLM input prompts. Our approach effectively reduces the number of input tokens by up to 1/6 compared to directly querying online LLMs, while consistently outperforming traditional NLP methods and other LLM-based strategies.

Figures

Figures reproduced from arXiv: 2505.12792 by the authors.

Figure 1
Figure 1. An illustration of (a) human value system and (b) the proposed EAVIT method compared with directly using LLMs. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. An illustration of the Schwartz value systems. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Value class distribution of original Touch [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Plot of F1-scores and class distribution on the Touche23-ValueEval test set over the labels by level. The grey bars show the label ´ distribution. Webis-ArgValues-22 [Kiesel et al., 2022], and Touche23- ´ ValueEval [Kiesel et al., 2023]. Details can be found in Ap￾pend…
Figure 5
Figure 5. Figure 5: Sample Visualization of Value Identification of Virtual Individuals [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Results of different models with different finetune strate [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Output variance at different output stages. [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Internal Pluralism and the Limits of Pairwise Comparisons

    cs.AI 2026-07 conditional novelty 7.0 of 10

    Under internal pluralism, forced local pairwise comparisons erase inseparable priorities and distort conflicted answers, while allowing indecision reports can sharply reduce queries needed to learn preference weights.

Reference graph

Works this paper leans on

41 extracted references · 27 canonical work pages · cited by 1 Pith paper

  1. [1]

    Using large language models to sim- ulate multiple humans and replicate human subject studies

    [Aher et al., 2023] Gati V Aher, Rosa I Arriaga, and Adam Tauman Kalai. Using large language models to sim- ulate multiple humans and replicate human subject studies. In International Conference on Machine Learning , pages 337–371. PMLR,

    work page 2023

  2. [7]

    Qlora: Efficient fine- tuning of quantized llms,

    [Dettmers et al., 2023] Tim Dettmers, Artidoro Pagnoni, Ari Holtzman, and Luke Zettlemoyer. Qlora: Efficient fine- tuning of quantized llms,

    work page 2023

  3. [8]

    BERT: Pre-training of deep bidirectional transformers for language understand- ing

    [Devlin et al., 2019] Jacob Devlin, Ming-Wei Chang, Ken- ton Lee, and Kristina Toutanova. BERT: Pre-training of deep bidirectional transformers for language understand- ing. pages 4171–4186, June

    work page 2019

  4. [10]

    Social chemistry 101: Learning to reason about social and moral norms

    [Forbes et al., 2020] Maxwell Forbes, Jena D Hwang, Vered Shwartz, Maarten Sap, and Yejin Choi. Social chemistry 101: Learning to reason about social and moral norms. arXiv preprint arXiv:2011.00620,

    arXiv 2020

  5. [11]

    Common morality: Deciding what to do

    [Gert, 2004] Bernard Gert. Common morality: Deciding what to do. Oxford University Press,

    work page 2004

  6. [14]

    Large language models can self-improve.arXiv preprint arXiv:2210.11610,

    [Huang et al., 2022] Jiaxin Huang, Shixiang Shane Gu, Le Hou, Yuexin Wu, Xuezhi Wang, Hongkun Yu, and Ji- awei Han. Large language models can self-improve.arXiv preprint arXiv:2210.11610,

    arXiv 2022

  7. [15]

    World values surveys and european values surveys, 1981- 1984, 1990-1993, and 1995-1997

    [Inglehart et al., 2000] Ronald Inglehart, Miguel Basanez, Jaime Diez-Medrano, Loek Halman, and Ruud Luijkx. World values surveys and european values surveys, 1981- 1984, 1990-1993, and 1995-1997. Ann Arbor-Michigan, Institute for Social Research, ICPSR version,

    work page 2000

  8. [18]

    Semeval- 2023 task 4: Valueeval: Identification of human values be- hind arguments

    [Kiesel et al., 2023] Johannes Kiesel, Milad Alshomary, Nailia Mirzakhmedova, Maximilian Heinrich, Nicolas Handke, Henning Wachsmuth, and Benno Stein. Semeval- 2023 task 4: Valueeval: Identification of human values be- hind arguments. In Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023), pages 2287–2303,

    work page 2023

Show all 41 references
  1. [19]

    Roberta: A robustly optimized bert pretraining approach

    [Liu et al., 2019] Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692,

    2019 arXiv

  2. [20]

    Lost in the middle: How language mod- els use long contexts

    [Liu et al., 2023] Nelson F Liu, Kevin Lin, John Hewitt, Ashwin Paranjape, Michele Bevilacqua, Fabio Petroni, and Percy Liang. Lost in the middle: How language mod- els use long contexts. arXiv preprint arXiv:2307.03172 ,

    2023 arXiv

  3. [21]

    Explanation-based finetuning makes models more robust to spurious cues

    [Ludan et al., 2023] Josh Magnus Ludan, Yixuan Meng, Tai Nguyen, Saurabh Shah, Qing Lyu, Marianna Apidianaki, and Chris Callison-Burch. Explanation-based finetuning makes models more robust to spurious cues. In Anna Rogers, Jordan Boyd-Graber, and Naoaki Okazaki, edi- tors, Pr...

    2023

  4. [22]

    [Meng et al., 2022] Yu Meng, Jiaxin Huang, Yu Zhang, and Jiawei Han

    Association for Computational Linguistics. [Meng et al., 2022] Yu Meng, Jiaxin Huang, Yu Zhang, and Jiawei Han. Generating training data with language mod- els: Towards zero-shot language understanding,

    2022

  5. [23]

    Tuning language models as training data generators for augmentation-enhanced few-shot learning

    [Meng et al., 2023] Yu Meng, Martin Michalski, Jiaxin Huang, Yu Zhang, Tarek Abdelzaher, and Jiawei Han. Tuning language models as training data generators for augmentation-enhanced few-shot learning. In Interna- tional Conference on Machine Learning , pages 24457– 24477. PMLR,

    2023

  6. [24]

    Who is gpt-3? an exploration of personality, values and demographics

    [Miotto et al., 2022] Maril`u Miotto, Nicola Rossberg, and Bennett Kleinberg. Who is gpt-3? an exploration of personality, values and demographics. arXiv preprint arXiv:2209.14338,

    2022 arXiv

  7. [25]

    Predictors of objective and subjective career success: A meta-analysis

    [Ng et al., 2005] Thomas WH Ng, Lillian T Eby, Kelly L Sorensen, and Daniel C Feldman. Predictors of objective and subjective career success: A meta-analysis. Personnel psychology, 58(2):367–408,

    2005

  8. [27]

    Discovering language model behav- iors with model-written evaluations

    [Perez et al., 2022] Ethan Perez, Sam Ringer, Kamil ˙e Lukoˇsi¯ut˙e, Karina Nguyen, Edwin Chen, Scott Heiner, Craig Pettit, Catherine Olsson, Sandipan Kundu, Saurav Kadavath, et al. Discovering language model behav- iors with model-written evaluations. arXiv preprint arXiv:2212.09251,

    2022 arXiv

  9. [28]

    Val- uenet: A new dataset for human value driven dialogue sys- tem

    [Qiu et al., 2022] Liang Qiu, Yizhou Zhao, Jinchao Li, Pan Lu, Baolin Peng, Jianfeng Gao, and Song-Chun Zhu. Val- uenet: A new dataset for human value driven dialogue sys- tem. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pages 11183–11191,

    2022

  10. [29]

    Valuebench: Towards comprehensively evaluating value orientations and un- derstanding of large language models

    [Ren et al., 2024] Yuanyi Ren, Haoran Ye, Hanjun Fang, Xin Zhang, and Guojie Song. Valuebench: Towards comprehensively evaluating value orientations and un- derstanding of large language models. arXiv preprint arXiv:2406.04214,

    2024 arXiv

  11. [32]

    Monitoring social values in online media articles on child vaccinations

    [Scharfbillig et al., 2022] Mario Scharfbillig, Vladimir Ponizovskiy, Zsuzsanna P´asztor, Julian Keimer, Giuseppe Tirone, et al. Monitoring social values in online media articles on child vaccinations. Technical report, Technical Report KJ-NA-31-324-EN-N, European Commission’s...

    2022

  12. [33]

    Adam-smith at semeval-2023 task 4: Discovering human values in arguments with en- sembles of transformer-based models

    [Schroter et al., 2023] Daniel Schroter, Daryna Dementieva, and Georg Groh. Adam-smith at semeval-2023 task 4: Discovering human values in arguments with en- sembles of transformer-based models. arXiv preprint arXiv:2305.08625,

    2023 arXiv

  13. [34]

    Extending the cross-cultural validity of the theory of basic human values with a different method of measure- ment

    [Schwartz et al., 2001] Shalom H Schwartz, Gila Melech, Arielle Lehmann, Steven Burgess, Mari Harris, and Vicki Owens. Extending the cross-cultural validity of the theory of basic human values with a different method of measure- ment. Journal of cross-cultural psychology , 32(...

    2001

  14. [37]

    Llama 2: Open foundation and fine-tuned chat models

    [Touvron et al., 2023b] Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288,

    arXiv

  15. [38]

    Self-instruct: Aligning language model with self generated instructions

    [Wang et al., 2022] Yizhong Wang, Yeganeh Kordi, Swa- roop Mishra, Alisa Liu, Noah A Smith, Daniel Khashabi, and Hannaneh Hajishirzi. Self-instruct: Aligning language model with self generated instructions. arXiv preprint arXiv:2212.10560,

    2022 arXiv

  16. [39]

    Chain-of-thought prompting elicits reasoning in large language models,

    [Wei et al., 2023] Jason Wei, Xuezhi Wang, Dale Schuur- mans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. Chain-of-thought prompting elicits reasoning in large language models,

    2023

  17. [40]

    Value fulcra: Mapping large lan- guage models to the multidimensional spectrum of basic human values

    [Yao et al., 2023] Jing Yao, Xiaoyuan Yi, Xiting Wang, Yi- fan Gong, and Xing Xie. Value fulcra: Mapping large lan- guage models to the multidimensional spectrum of basic human values. arXiv preprint arXiv:2311.10766,

    2023 arXiv

  18. [41]

    Be creative

    8 Appendix 8.1 Prompt Templates Prompt Template for Generating Explanation // Simplified prompt template ... You are an expert on Schwartz Theory of Basic Values. Below is the definition of human value [ VALUE]: [DEFINITION] For the following text, can you explain how the text...

    2022

  19. [1973]

    Values and identities-a policy- maker’s guide

    [Scharfbillig et al., 2021] Mario Scharfbillig, Laura Smillie, David Mair, Marta Sienkiewicz, Julian Keimer, R Pinho Dos Santos, H Vinagreiro Alves, Elisa Vecchione, and Laurenz Scheunemann. Values and identities-a policy- maker’s guide. Publications Office of the European Union,

    2021

  20. [2000]

    World values survey: Round six - country-pooled datafile version: https://www.worldvaluessurvey.org/wvsdocumentationwv6.jsp.,

    [Inglehart et al., 2014] Ronald Inglehart, Miguel Basanez, Jaime Diez-Medrano, Loek Hal- man, and Ruud Luijkx. World values survey: Round six - country-pooled datafile version: https://www.worldvaluessurvey.org/wvsdocumentationwv6.jsp.,

    2014

  21. [2001]

    An overview of the schwartz theory of basic values

    [Schwartz, 2012] Shalom H Schwartz. An overview of the schwartz theory of basic values. Online readings in Psy- chology and Culture, 2(1):11,

    2012

  22. [2004]

    Moral foundations theory: The pragmatic validity of moral pluralism

    [Graham et al., 2013] Jesse Graham, Jonathan Haidt, Sena Koleva, Matt Motyl, Ravi Iyer, Sean P Wojcik, and Peter H Ditto. Moral foundations theory: The pragmatic validity of moral pluralism. In Advances in experimental social psychology, volume 47, pages 55–130. Elsevier,

    2013

  23. [2005]

    Gpt-4 technical report

    [OpenAI, 2023] OpenAI. Gpt-4 technical report. arXiv preprint arXiv:2303.08774,

    2023 arXiv

  24. [2010]

    Basic human val- ues and moral foundations theory in valuenet ontology

    [De Giorgis et al., 2022] Stefano De Giorgis, Aldo Gangemi, and Rossana Damiano. Basic human val- ues and moral foundations theory in valuenet ontology. In International Conference on Knowledge Engineering and Knowledge Management, pages 3–18. Springer,

    2022

  25. [2012]

    Hashimoto

    [Taori et al., 2023] Rohan Taori, Ishaan Gulrajani, Tianyi Zhang, Yann Dubois, Xuechen Li, Carlos Guestrin, Percy Liang, and Tatsunori B. Hashimoto. Stanford alpaca: An instruction-following llama model. https://github.com/ tatsu-lab/stanford alpaca,

    2023

  26. [2013]

    Lora: Low-rank adaptation of large language models

    [Hu et al., 2021] Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685,

    2021 arXiv

  27. [2014]

    Identifying the Human Values behind Argu- ments

    [Kiesel et al., 2022] Johannes Kiesel, Milad Alshomary, Nicolas Handke, Xiaoni Cai, Henning Wachsmuth, and Benno Stein. Identifying the Human Values behind Argu- ments. In Smaranda Muresan, Preslav Nakov, and Aline Villavicencio, editors, 60th Annual Meeting of the Asso- ciati...

    2022

  28. [2019]

    What does chatgpt return about human values? exploring value bias in chatgpt using a descriptive value theory

    [Fischer et al., 2023] Ronald Fischer, Markus Luczak- Roesch, and Johannes A Karl. What does chatgpt return about human values? exploring value bias in chatgpt using a descriptive value theory. arXiv preprint arXiv:2304.03612,

    2023 arXiv

  29. [2020]

    Developing a meta-inventory of hu- man values

    [Cheng and Fleischmann, 2010] An-Shou Cheng and Ken- neth R Fleischmann. Developing a meta-inventory of hu- man values. Proceedings of the American Society for In- formation Science and Technology, 47(1):1–10,

    2010

  30. [2021]

    Language models are few-shot learners

    [Brown et al., 2020] Tom Brown, Benjamin Mann, Nick Ry- der, Melanie Subbiah, Jared D Kaplan, Prafulla Dhari- wal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. Language models are few-shot learners. Advances in neural information processing sys- tems,...

    2020

  31. [2022]

    [Atkinson and Bench-Capon, 2021] Katie Atkinson and Trevor Bench-Capon

    Association for Computational Linguistics. [Atkinson and Bench-Capon, 2021] Katie Atkinson and Trevor Bench-Capon. Value-based argumentation. Journal of Applied Logics, 8(6):1543–1588,

    2021

  32. [2023]

    The moral debater: A study on the computational generation of morally framed arguments

    [Alshomary et al., 2022] Milad Alshomary, Roxanne El Baff, Timon Gurcke, and Henning Wachsmuth. The moral debater: A study on the computational generation of morally framed arguments. In Smaranda Muresan, Preslav Nakov, and Aline Villavicencio, editors, Proceedings of the 60th...

    2022

  33. [2024]

    The nature of human val- ues

    [Rokeach, 1973] Milton Rokeach. The nature of human val- ues. Free press,

    1973

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.