cs.AIOct 15, 2026
SaveEntertain, then Cancel: How Thinking Makes Reasoning Models Read Some and Or Literally
Organizations: 1The Hong Kong University of Science and Technology · 2The Hong Kong University of Science and Technology (Guangzhou) · Peking University
Abstract
When a speaker says ``some students passed'', listeners infer that not all of them did. Reasoning models now let users decide how long a model thinks before it answers, but we do not know what extra thinking does to such everyday inferences. We study this directly: we keep a model's weights fixed, change only how much it thinks, and record which reading it chooses. A second tool cuts the model's own reasoning at successive points and asks for an answer, showing where the reading changes. Thinking makes models read logical words as logic: it cancels the implicatures of some (not all) and or (not both) in most tested open reasoning models and in GPT-5, while context-driven inferences such as indirect answers improve. The trace shows why: models first entertain the enriched reading and cancel it later. Answering from that early reading, an exploratory inference-time remedy, recovers these implicatures without losing literal accuracy. Evaluations of reasoning models should measure how they interpret, not only how accurate they are.
Figures & tables
Figure 1: Thinking acts as a logic lens on logical words. some and or are read as and , while adequate keeps ‘not good’.
Figure 2: Thinking makes some more literal in most open reasoning models. Vermilion: counted literal shift; blue: more pragmatic ( Figure 11 ).
Figure 3: A reading probe on fixed weights. Items with a trigger (①) are answered with thinking off, capped or natural (②); we read the chosen reading and the log-odds (③), also after successive trace prefixes (④), for every model and checkpoint.
| Logical words | Rest of the battery | Three-way choice | |||||
| Model | some | or | modals | lexical scales | indirect ans. | some | or |
| Qwen3 (thinking off natural thinking) | |||||||
| Qwen3-4B | 37.2 [ 55.1, 17.9] | 61.6 [ 74.7, 47.5] | 32.3 [ 47.9, 15.6] | 7.7 [ 11.9, 4.0] | 0.4 [ 7.2, 6.8] ∘ | 9.9 | 11.9 |
| Qwen3-8B | 28.2 [ 41.0, 16.7] | 36.4 [ 49.5, 24.2] | 12.5 [ 22.9, 2.1] | 5.5 [2.4, 8.8] | 6.5 [ 1.1, 14.7] ∘ | 34.0 | 13.8 |
| Qwen3-14B | 21.8 [ 37.2, 9.0] | 13.1 [ 21.2, 6.1] | 4.2 [ 3.1, 13.5] ∘ | 6.8 [3.3, 10.8] | 1.4 [ 5.4, 8.2] ∘ | 11.4 | 2.4 |
| Qwen3-30B-A3B | 28.2 [ 39.7, 17.9] | 39.4 [ 51.5, 26.3] | 19.8 [ 36.5, 2.1] | 0.7 [ 2.6, 4.0] ∘ | 9.5 [2.5, 16.8] | 12.8 | 11.8 |
Table 1: Thinking reduces the pragmatic reading of logical words across families. Change with thinking (pp) [95% CI]; bold / underlined : largest/second absolute change per column; circ : CI includes 0.
Figure 4: The enriched reading is endorsed early in the trace and abandoned later. Pragmatic reading after the first tokens of the model’s own trace, with 95% bands; panel (d) and the lexical-scale lines come from an earlier, smaller run of the probe.
| Analysis choice | some | or |
| Qwen3-8B | ||
| Reference: core, Yes/No, off natural | 28.2 | 36.4 |
| Baseline: forced-empty thinking block | 39.7 | 33.3 |
| Budget 128, stop phrase | 37.2 | 4.0 |
| Budget 512, stop phrase | 5.1 | 15.2 |
| Budget 1024, stop phrase | 21.8 | 31.3 |
Table 2: Full thinking lowers the reading under every analysis choice. Change in the pragmatic reading with thinking (pp); short budgets raise it before it falls.
Figure 5: Pragmatic reading rate for Qwen3-8B vs. thinking budget. Capped with stop phrase (hollow: plain cap). Lower strip: realized tokens. Bands: 95% cluster-bootstrap CIs.
| Default reading (%) | ||||
| Stage | some | or | lexical | LUDWIG |
| OLMo-3-Think (native thinking; cost = NAT forced-empty) | ||||
| base | 75.6 4.4 | 74.7 1.8 | 39.5 0.4 | 67.0 4.0 |
| Think-SFT | 39.7 5.9 | 50.5 3.5 | 5.3 0.7 | 76.8 2.8 |
| Think-DPO | 53.8 3.8 | 48.5 5.2 | 4.6 0.7 | 75.1 2.4 |
| RL step 100 | 47.4 2.2 | 51.5 0.0 | 5.3 0.7 | 75.1 3.4 |
Table 3: The literal default appears with thinking SFT in OLMo-3 and DPO in Tülu-3. Default reading (%) under a fixed 4-shot prompt, mean s.d..
| Model | Method | Felicity | Speaker | Presup. | Literal |
| Qwen3-8B | Natural thinking | 1.2 | 58.2 | 65.7 | 87.9 |
| Task-aware instruction | 3.5 | 60.2 | 66.8 | 92.0 | |
| ours | 20.5 | 62.5 | 65.7 | 87.9 | |
| DeepSeek-R1 | Natural thinking | 2.2 | 55.5 | 62.3 | 86.3 |
| Task-aware instruction | 2.6 | 57.8 | 62.2 | 87.0 | |
| ours | 8.1 | 58.8 | 62.9 | 86.3 |
Table 4: Answering from the entertained reading restores scalar implicatures without losing literal accuracy (held-out test families; accuracy, %). Felicity: IMPPRES some / or felicity items.
Appendix figures & tables20 assets
Supplementary material from the paper’s appendix.
Appendix
| Group | Hugging Face identifiers |
| Hybrid thinking | Qwen/Qwen3-{0.6B,1.7B,4B,8B,14B,30B-A3B,32B} , Qwen/Qwen3-8B-Base ( Yang et al., 2025 ) |
| Effort levels | openai/gpt-oss-{20b,120b} ( OpenAI et al., 2025 ) |
| OLMo-2 stages | allenai/OLMo-2-1124-7B , -SFT , -DPO , -Instruct ; 13B analogues ( Team OLMo et al., 2024 ) |
| OLMo-3 stages | allenai/Olmo-3-1025-7B , Olmo-3-7B-Think-SFT , -Think-DPO , -Think ; RL-Zero {Math, Code, IF, General, Mix} step branches ( Team Olmo et al., 2025 ) |
| Tülu-3 stages | allenai/Llama-3.1-Tulu-3-8B-SFT , -DPO , Llama-3.1-Tulu-3-8B , Llama-3.1-Tulu-3.1-8B ( Lambert et al., 2024 ) |
| Distilled / zero-RL | deepseek-ai/DeepSeek-R1-Distill-Qwen-{1.5B,7B,14B} ( DeepSeek-AI, 2025 ) ; Qwen2.5 bases ( Qwen et al., 2024 ) ; SimpleRL-Zoo ( Zeng et al., 2025 ) ; Open-Reasoner-Zero ( Hu et al., 2025 ) |
Table 5: Models by group. Commit hashes of every checkpoint are recorded with each run.
| Measure | nt | 512 tokens | nat |
| LUDWIG, prompts of Ruis et al. (2023) : accuracy | |||
| 0-shot | 79.7 | 84.0 | 82.0 |
| 5-shot | 71.9 | 79.6 | 78.6 |
| 30-shot | 67.6 | 78.1 | 78.7 |
| ImplicatureX ( Spinoso-Di Piano et al., 2026 ) : accuracy | |||
| Scalar, recognition | 100.0 | 89.9 | 65.2 |
Table 6: On ImplicatureX, thinking lowers scalar-implicature recognition and raises cancellation, as Spinoso-Di Piano et al. (2026) report. Qwen3-8B, accuracy (%) without thinking, with a 512-token cap and with natural thinking.
| Pragmatic reading | Yes-bias checks | ||||
| Model | Dose contrast | Yes/No inference | Three-way choice | “Yes” without utterance | Pragmatic, flipped question |
| GPT-5 | minimal high | 15.2 * | 8.8 * | 26.9 * | 3.4 |
| Claude Sonnet 4.6 | off 16k | 19.2 * | 9.7 * | 7.9 * | 13.0 * |
| Claude Opus 4.8 | off high | 10.8 * | 1.3 | 3.9 | 5.0 * |
Table 7: Closed models: robustness of the thinking effect on logical-word items (pp, highest minus lowest dose), on items answered validly at both doses; two samples per cell; ∗ : paired 95% cluster-bootstrap CI excludes 0. “Yes” without utterance: change in the rate of answering Yes when the speaker’s sentence is removed. Flipped question: change in the pragmatic rate when the question is reworded so that No is the pragmatic answer. A pure shift towards Yes raises the first and lowers the second. In the three-way format pragmatic answers are rare for every model (0–9%), which limits how far the rate can rise. Claude baselines use a 512-token answer cap.
| Logical words | Lexical scales | Context | |||||
| Model | Dose contrast | some | or | modals | gradable adj. | scales | LUDWIG |
| GPT-5 | minimal high effort | 20.6 4.4 * | 18.8 10.6 * | 12.5 3.5 * | 0.0 3.5 | 7.5 1.2 * | 6.2 1.8 |
| GPT-5.4 | minimal high effort | 2.5 | 20.0 * | 15.0 * | 7.5 | 15.0 * | 2.5 |
| o4-mini | low high effort | 1.2 | 6.2 | 12.5 | 5.0 | 5.0 | 2.5 |
| DeepSeek-V4-Pro ¶ | thinking off on | 8.9 3.6 * | 1.9 6.3 | 1.2 1.8 | 8.8 4.8 * | 41.7 2.2 * | 10.0 3.5 * |
| DeepSeek-V4-Flash ¶ | thinking off on | 16.9 2.7 * | 4.4 2.7 | 1.2 1.8 | 5.0 3.5 | 10.8 1.2 * | 5.0 0.0 |
Table 8: Closed and API-served models: change in the pragmatic reading (pp) from the lowest to the highest available thinking dose; LUDWIG is accuracy. ∗ : 95% cluster-bootstrap CI over items excludes 0. ¶ Open-weight DeepSeek-V4 served through our LiteLLM proxy; Yes/No format only. ‡ Gemini 2.5 Pro cannot switch thinking off; 128 tokens is its minimum budget. Robustness checks in Table 7 ; 80 some , 80 or and 40 modal items.
| Yes/No inference | Three-way choice | |||
| Model | some | or | some | or |
| Qwen3-4B | 49.8 [ 53.9, 45.6] | 53.0 [ 57.6, 48.4] | 9.9 [ 12.7, 7.3] | 11.9 [ 14.7, 9.2] |
| Qwen3-8B | 37.7 [ 42.1, 33.3] | 37.7 [ 41.6, 33.8] | 34.0 [ 38.2, 29.9] | 13.8 [ 17.1, 10.6] |
| Qwen3-14B | 22.6 [ 27.1, 18.3] | 15.9 [ 19.1, 12.7] | 11.4 [ 14.4, 8.6] | 2.4 [ 4.0, 1.0] |
| Qwen3-30B-A3B | 35.7 [ 39.8, 31.4] | 40.9 [ 45.1, 36.8] | 12.8 [ 15.7, 10.1] | 11.8 [ 14.3, 9.3] |
| Qwen3-32B | 55.6 [ 60.1, 50.8] | 52.7 [ 57.0, 48.3] | 38.0 [ 42.9, 33.1] | 19.7 [ 23.1, 16.3] |
Table 9: On the full held-out sets, some and or turn literal with thinking in both formats in most models. Change with thinking (pp) [95% CI], 300 items per trigger; grey: CI includes 0. Some models start near the floor in the three-way format.
Figure 6: Trace probe on further families (EXAONE-Deep-7.8B, Nemotron-Nano-9B-v2, Phi-4-mini-reasoning, Skywork-OR1-7B, Llama-3.1-Nemotron-Nano-8B): pragmatic reading after the first tokens of the model’s own trace, with 95% bands over items. The -axis runs from 0 to 100 (0 to 80 in Figure 4 ). In panel (e) the value at is the model’s collapse to one answer under a forced-empty thinking block.
Figure 7: Trace probe on the newest generation (Qwen3.5-4B, Qwen3.5-9B, Qwen3.8-27B): pragmatic reading after the first tokens of the model’s own trace, with 95% bands over items. The early lexical-scale jump in the Qwen3.5 panels is systematic and confined to sampled answers (see text).
| Model | Circa | NOPE |
| Qwen3-4B | 15.3 [10.4, 20.4] | 7.8 [2.3, 13.1] |
| Qwen3-8B | 17.5 [12.4, 22.7] | 9.9 [4.5, 15.5] |
| Qwen3-14B | 24.4 [18.9, 30.0] | 9.6 [4.4, 14.9] |
| Qwen3-30B-A3B | 8.3 [4.0, 12.7] | 10.4 [5.3, 15.5] |
| Qwen3-32B | 3.1 [ 0.9, 7.1] | 9.2 [3.8, 14.4] |
| Qwen3.5-4B | 16.7 [12.9, 20.4] | 5.8 [1.4, 10.4] |
Table 10: Thinking raises or keeps accuracy on two further context-driven sets in every model. Change with thinking (pp) [95% CI]; grey: CI includes 0; Circa ( Louis et al., 2020 ) , NOPE ( Parrish et al., 2021 ) .
Figure 8: Thinking effect for every item family and model. Change in pragmatic reading with thinking (pp); vermilion: more literal; blue: more pragmatic; framed: 95% CI excludes 0; hatched: not run. Values in Tables 12 and 8 .
| Qwen3-8B | Qwen3-32B | |||||
| Method | some | or | LUDWIG | some | or | LUDWIG |
| No thinking ( nt ) | 29.5 5.9 | 37.4 3.5 | 76.5 2.6 | 65.4 3.8 | 55.6 3.5 | 83.9 1.2 |
| Natural thinking ( nat ) | 0.0 0.0 | 1.0 1.8 | 83.5 1.6 | 15.4 3.8 | 8.1 3.5 | 83.5 1.6 |
| Prompted CoT, thinking off a | 10.3 5.9 | 6.1 3.0 | 75.4 2.2 | 46.2 10.2 | 45.5 13.9 | 80.7 2.2 |
| Seed sentence opening the thinking trace (26 tokens each) | ||||||
| Neutral seed | 2.6 2.2 | 1.0 1.8 | 77.5 2.2 | 10.3 5.9 | 6.1 6.1 | 84.9 2.2 |
Table 11: The intent seed restores more enrichment than length-matched control seeds. Pragmatic reading and LUDWIG accuracy (%); mean s.d. over seeds; bold / underlined : best/second among thinking rows.
Figure 9: Thinking makes some and or more literal in every major model family. Change with thinking (pp, 95% CI). Size trends: Figure 10 .
Figure 10: The literal shift holds across sizes and families. Thinking effect on some and or (pp, 95% CI) by model size for Qwen3, Qwen3.5 and R1-Distill-Qwen, with Qwen3.8, MiMo, openPangu and DeepSeek-R1-0528 as single points.
| Generalized implicatures: (pp) | Acc. (pp) | |||||
| Model | some | or | modals | lexical scales | IMPPRES all | LUDWIG |
| Qwen3, distilled from larger models (0.6B–30B-A3B) or fully RL-trained (32B); nat nt | ||||||
| Qwen3-0.6B † | 50.0 10.2 * | 83.8 4.6 * | 62.5 12.5 * | 70.4 2.0 * | 59.1 1.6 * | 9.8 3.2 * |
| Qwen3-1.7B | 10.3 8.9 | 47.5 4.6 * | 33.3 9.5 * | 25.4 7.3 * | 23.6 1.8 * | 17.2 0.6 * |
| Qwen3-4B | 37.2 4.4 * | 61.6 4.6 * | 32.3 14.4 * | 7.7 2.0 * | 26.5 2.8 * | 0.4 2.7 |
| Qwen3-8B | 28.2 5.9 * | 36.4 5.2 * | 12.5 6.2 * | 5.5 1.4 * | 9.9 2.4 * | 6.5 2.8 |
Table 12: Full results: the literal shift on some and or holds in 30 of 40 open reasoning models. Change in the reading chosen from no thinking ( nt ) to natural thinking ( nat ) on the same weights, by item family, on the held-out split; for OLMo-3-Think the baseline is a forced-empty thinking block (the model has no switch), and for gpt-oss the contrast is high minus low reasoning effort. Generalized families use the inference framing: IMPPRES quantifier, connective and modal triggers ( Jeretic et al., 2020 ) and scalar-diversity lexical scales ( van Tiel et al., 2016 ; Gotzner et al., 2018 ; Pankratz and van Tiel, 2021 ) . LUDWIG: indirect answers ( Ruis et al., 2023 ) . Cells: mean standard deviation over 3 decoding seeds; blue = more pragmatic or more accurate, vermilion = more literal; * : the 95% cluster-bootstrap confidence interval over items excludes 0 (uncorrected; the mixed-effects tests are in Section C.1 ). † Without thinking, Qwen3-0.6B accepts 86–100% of inferences (a yes-bias), so it is outside the pre-registered 10–90% range and is excluded from the dose–response analysis. Table 14 lists the absolute rates.
| Hyp. | Prediction | Outcome (evidence) | Status |
| H1 effort | Pragmatic reading on generalized implicatures rises with thinking | Falls for some or or (CI excluding 0) in 30 of 40 open models, among them EXAONE-Deep-7.8B, Phi-4-mini-reasoning, every newest-generation model (Qwen3.5 0.8B–9B, Qwen3.8-27B), QwQ-32B, Nemotron-Super-49B, R1-Distill-Llama-70B, every Qwen3 size from 1.7B, gpt-oss-20b, OLMo-3-32B-Think, Phi-4-reasoning(-plus), R1-Distill-Llama-8B, MiMo, Nemotron-Nano, AceReason and Skywork-OR1; rises in Magistral-Small (a shift towards answering Yes , Section B.3 ) Granite-3.3, Llama-3.1-Nemotron-Nano-8B (Yes/No format only; literal in the three-way choice) and openPangu-Embedded-7B-model (Yes/No format only; flat in the three-way choice); no significant change in gpt-oss-120b, OLMo-3-7B-Think, SmolLM3 and openPangu-Embedded-7B-V1.1 ( Table 12 ); GPT-5 falls in two formats ( Section B.3 ) | not supported |
| H1 default-and-cancel | Falls under the felicity framings, 5 pp in 2 of 4B/8B/14B | Holds for some / or per family; the pre-registered IMPPRES aggregate at 14B is flat (0.5 pp), so the aggregate gate criterion failed | partly supported |
| H1b capability crossover | Negative at small sizes, positive at 32B on 2 families, reproduced in a same-recipe pair | Qwen3-32B negative on some / or ; OLMo-3-Think 7B null and 32B positive on IMPPRES but negative on some ; no family flips sign within a pair on 2 families | not supported |
| H2 split | PCIs improve while GCIs fall; 3 PCI sets improve or stay flat | Accuracy rises on Fig-QA and PUB at every Qwen3 size from 4B to 32B (8B: +10.7 and +10.1 pp), on Circa and NOPE in 14 models with no decrease ( Table 10 ), and on presupposition and LUDWIG in Qwen3-8B; the budget class interaction is significant in Qwen3-4B, 8B and 14B ( Section D.3 ) | supported |
| H2b speaker knowledge | Default-and-cancel: thinking lowers in both knowledge conditions; effort/capability: thinking widens the gap | Knowledge sensitivity 0 in Qwen3 and OLMo; in gpt-oss-120b it shrinks with effort (84.0 62.5 pp) | default-and-cancel supported (one model) |
| H2c framing class | Truth framing denies implicatures of both classes; meaning framing lets PCIs improve | the interpretation-truth arm was not run | not tested |
Table 13: Outcome of every pre-registered prediction (held-out split). “Not tested” marks predictions whose arm was not run.
| Qwen3-4B | Qwen3-8B | Qwen3-14B | |||||||
| Phenomenon | nt | nat | nt | nat | |||||
| Generalized implicatures: pragmatic reading (%) | |||||||||
| Category sentences, ternary | 5.0 0.0 | 2.5 0.0 | 2.5 | 2.5 0.0 | 0.0 0.0 | 2.5 | 2.5 0.0 | 0.0 0.0 | 2.5 |
| Scalar diversity | 16.6 1.1 | 8.8 1.0 | 7.7 | 4.9 0.4 | 10.4 1.0 | 5.5 | 2.0 0.7 | 8.8 1.4 | 6.8 |
| IMPPRES scalar | 49.7 1.8 | 23.2 1.3 | 26.5 | 22.6 1.8 | 12.7 1.2 | 9.9 | 14.5 1.5 | 15.0 2.1 | 0.5 |
| Presupposition: projection (%) | |||||||||
Table 14: Per-phenomenon results for Qwen3-4B, 8B and 14B. Reading chosen without ( nt ) and with natural thinking ( nat ), mean s.d. over three seeds, held-out split; in pp; –: not run in this form.
| Hyp. | Prediction | Outcome (evidence) | Status |
| Scenario validity | Humans find some misleading only under full access ( 80%) | 100.0% of 24 (blind consensus labels) | met |
| Open-answer coding | LLM reference rater agrees with the human audit (lower bound of ) | best rater = 0.57 [0.34, 0.79]; open answers reported descriptively | not met |
| Bott–Noveck items | Category sentences move with thinking | Near the literal ceiling in every format and condition for Qwen3 4B–32B; only binary felicity at 14B moves | floor set |
Table 13 , continued (part 3 of 3).
| Hyp. | Prediction | Outcome (evidence) | Status |
| H3 training origin | The literal shift appears at RLVR, not SFT; grows with RL steps | Tülu-3: at DPO, before RLVR; OLMo-3: at thinking SFT, RL-Zero flat, and the thinking cost shrinks over RL steps ( Table 3 ) | mixed |
| H4 load | A load lowers without thinking | With 2,048 words of unrelated preceding text and thinking off, on logical words moves from 22.6 to 14.2% in Qwen3-8B but from 14.5 to 20.9% in Qwen3-14B; no consistent direction | not supported |
| H5 human ordering | with human rates per scale | not run | not tested |
| H6 restorable | IST restores 50% of the shift; a length-matched neutral seed does not | With 26-token seeds, the intent seed recovers 12.8% of a 29.5% no-thinking reading on some in Qwen3-8B ( nat : 0.0%); neutral and logic seeds recover little or none ( Table 11 ) | partly met: specific to intent, below 50% |
| H7 locus | The literal preference appears inside the trace, after a marker | Truncating the model’s own trace: the pragmatic reading of some first rises (Qwen3-8B, +23.1 pp after 128 tokens) and falls below the no-thinking level later in the trace ( Figure 4 ); literalness markers such as technically precede the change in only 20.6% of changing Qwen3-8B traces, and occur as often in traces that do not change (39.1%) | crossing supported; marker order not supported |
Table 13 , continued (part 2 of 3).
Figure 11: Thinking reduces enrichment of some , or or both in most tested models; other inference types move differently. Change in pragmatic reading with thinking (pp, 95% CI): vermilion, mean of some and or ; blue, mean of lexical scales and indirect answers; , three-way choice. Every family: Figure 8 . Q3.5-35B: Qwen3.5-35B-A3B.
| Prediction | Measurement on LMs | Hyp. |
| Effort ordering | on GCIs vs. under felicity framings (inference; ternary and 1–7 for category sentences); true/false as the logic-mode contrast | Section D.1 |
| GCI/PCI split | budget class interaction on (PCI: accuracy, literal-error share) | Section D.1 |
| Speaker knowledge | (full partial knowledge) at nt vs. nat | H2b |
| Framing class | PCI accuracy under meaning vs. interpretation-truth framing, nt vs. nat | H2c |
| Training origin | default by stage (log-prob scoring); cost of thinking by stage; (RL step, literal rate) | Section D.1 |
| Load | under a memory load, crossed with thinking | Section D.1 |
Table 15: From theory to measurement. Each prediction that separates the effort and default-and-cancel accounts, the quantity we measure, and the hypothesis it tests. : proportion of committed answers that take the pragmatic reading; : thinking budget in tokens.
| Block | Models | Items | Seeds | Cap | |
| E1 | 3 | 3,877 | 3 | 0.6 | 8192 |
| E1S | 13 | 2,052 | 3 | 0.6 | 8192 |
| E1SF | 5 | 3,600 | 3 | 0.6 | 8192 |
| E1SFX | 17 | 3,600 | 3 | 0.6 | 8192 |
| E1Q | 6 | 2,052 | 3 | 0.6 | 8192 |
| E1P | 7 | 2,052 | 3 | 0.6 | 8192 |
Table 16: Run settings per experiment block: models, items per model, decoding seeds, temperature and thinking cap (tokens).
More in this topic
Pragmatic Reasoning in Language Models
Pragmatic reasoning in language models studies how models infer and communicate meanings shaped by context, intent, shared knowledge, and social norms.
See all 103 papers