Paper Citation Record · LEDGER

Training language models to be warm and empathetic makes them less reliable and more sycophantic

As of 18 August 2026, this Paper Citation Record lists 47 of 47 outbound references and 11 inbound Pith citation observations for arXiv:2507.21919.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2507.21919 v2

Coverage vector

measured 47 of 47 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-06T12:18:40.921302Z

measured 58 of 58 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-17T06:30:58.91139+00:00

measured 11 of 11 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-06T11:48:35.902247Z

measured 1 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: arxiv_reference, observed 2026-08-05T02:28:24.338817Z

Reference resolution

47 of 47 outbound references displayed

verified exact1
verified fuzzy19
unresolved27
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

2
arxiv_reference, observed 2026-08-05T02:28:24.338817Z

Outbound references

Observation 816a64db-ef7d-41bc-9761-436e5a8b6a19 · outbound

This paper cites et al.Multi-turn evaluation of anthropomorphic behaviours in large language models.

Training language models to be warm and empathetic makes them less reliable and more sycophantic et al.Multi-turn evaluation of anthropomorphic behaviours in large language models

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:35.913728Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 31f4e5d1-7ed8-4733-8d31-9b0fdca96eb2 · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 2

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:48.003421Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation f3fc96b1-6364-4c87-9881-72d3f3632231 · outbound

This paper cites OpenAI Model Spec (2025).

Training language models to be warm and empathetic makes them less reliable and more sycophantic OpenAI Model Spec (2025)

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:47.919507Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 6d336fd1-a3b0-455b-a66b-f49bf93e814b · outbound

This paper cites Claude’s Character (2024).

Training language models to be warm and empathetic makes them less reliable and more sycophantic Claude’s Character (2024)

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:47.796163Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation eb8ddd6b-6869-4127-bd1e-85d8e121ef49 · outbound

This paper cites Meet my A.I.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Meet my A.I

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:47.675259Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 7bdf4e34-7c8d-49dd-9f4d-f2047eebc879 · outbound

This paper cites Kids are in crisis.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Kids are in crisis

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:47.532498Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation ebb0f3cb-ec00-4955-b68d-2a48f16e8bca · outbound

This paper cites How people are really using Gen AI in 2025.

Training language models to be warm and empathetic makes them less reliable and more sycophantic How people are really using Gen AI in 2025

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:47.366409Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation f93c4fa9-a306-445c-b7a7-98e915c60301 · outbound

This paper cites R., Gabriel, I., Summerfield, C., Vidgen, B.

Training language models to be warm and empathetic makes them less reliable and more sycophantic R., Gabriel, I., Summerfield, C., Vidgen, B

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:47.275526Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 457b5d5b-53d7-42c1-bdc1-918b69d21454 · outbound

This paper cites (Im) Politeness, face and perceptions of rapport: unpackaging their bases and interrelationships.

Training language models to be warm and empathetic makes them less reliable and more sycophantic (Im) Politeness, face and perceptions of rapport: unpackaging their bases and interrelationships

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:47.181509Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 35090695-19ca-4af6-9a3a-92195fa001b8 · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 10

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:47.023722Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 3b49368a-4e84-448d-b5dc-2df53827807e · outbound

This paper cites & Gneezy, U.

Training language models to be warm and empathetic makes them less reliable and more sycophantic & Gneezy, U

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:46.857808Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 7b14e594-8d51-4479-af48-8d1cb4477e01 · outbound

This paper cites & Gabriel, I.

Training language models to be warm and empathetic makes them less reliable and more sycophantic & Gabriel, I

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:46.578040Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 87b4b85e-4229-4bba-991d-cb8ac2f6dacf · outbound

This paper cites Why AI chatbots lie to us.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Why AI chatbots lie to us

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:46.345657Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 75e73fcf-0bcb-4ab4-977d-260dbbbf939c · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 14

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:46.106036Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 20318753-f6dc-41fb-9ed0-f6500ee20a3f · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 15

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:45.913436Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation daeb99ae-b0ef-4f84-8b14-809ecbac30fb · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 16

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:45.661556Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 64f5a77d-9cd4-4a2f-a424-14688dfe2db3 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Training Verifiers to Solve Math Word Problems

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:37.817982Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c4fdc14f-445e-4781-9f32-f9361afa7df3 · outbound

This paper cites Universal and Transferable Adversarial Attacks on Aligned Language Models.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Universal and Transferable Adversarial Attacks on Aligned Language Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:37.932831Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d22da7be-8165-4b14-a5a1-9260e7623b47 · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 19

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:45.432053Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 25120187-ce11-47a6-ac71-2973f470603e · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 20

Resolution: verified exact; raw_fallback, observed 2026-08-06T12:18:41.391204Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 447b9046-0975-409c-9863-d245d58babca · outbound

This paper cites Lawsuit claims character.ai is responsible for teen’s suicide.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Lawsuit claims character.ai is responsible for teen’s suicide

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:45.194940Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation b42fed6d-c6b1-48cb-b151-958770101e94 · outbound

This paper cites They asked chatgpt questions.

Training language models to be warm and empathetic makes them less reliable and more sycophantic They asked chatgpt questions

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:44.942074Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation d076550f-7dea-4059-8393-6abddde7433f · outbound

This paper cites Character training: Understanding and crafting a language model’s personality (2025).

Training language models to be warm and empathetic makes them less reliable and more sycophantic Character training: Understanding and crafting a language model’s personality (2025)

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:44.698586Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation b1f2b7e7-1d28-49e0-875d-b71c662b8cd6 · outbound

This paper cites Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:38.621280Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8543145b-c384-4622-94a1-b819bc17969c · outbound

This paper cites Constitutional AI: Harmlessness from AI Feedback.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Constitutional AI: Harmlessness from AI Feedback

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:38.702350Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 240bfacc-8772-401e-93c7-1e5096ddd6f3 · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:38.810483Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 07c30131-6df1-4880-a999-ee67a68ff8d3 · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 27

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:44.498754Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 660d7f31-e222-4592-af6e-7761da6b7d9f · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 28

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:44.239773Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 8db69442-4a1b-4458-a6b2-a9aa6efa9713 · outbound

This paper cites et al.Accurate predictions on small data with a tabular foundation model.

Training language models to be warm and empathetic makes them less reliable and more sycophantic et al.Accurate predictions on small data with a tabular foundation model

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:44.010059Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation b5729b78-0700-4181-b3a1-5130d03b3cc7 · outbound

This paper cites & Jurafsky, D.

Training language models to be warm and empathetic makes them less reliable and more sycophantic & Jurafsky, D

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:43.788710Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation f94e2fab-c2ad-48f9-94ec-1794aed310c2 · outbound

This paper cites When Scaling Meets LLM Finetuning: The Effect of Data, Model and Finetuning Method.

Training language models to be warm and empathetic makes them less reliable and more sycophantic When Scaling Meets LLM Finetuning: The Effect of Data, Model and Finetuning Method

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:39.305805Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7244e46-3613-4554-9d43-5fe43cec5b8b · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 32

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:43.593434Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 67510a8d-1cb9-4898-b026-944fa97355b4 · outbound

This paper cites & Evans, O.

Training language models to be warm and empathetic makes them less reliable and more sycophantic & Evans, O

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:43.344806Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 34dcdba2-62e8-4d93-85ff-b8b3dfdb0997 · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:39.604278Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7d4633ec-6853-46fe-919f-44ba82fa97ed · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:39.748645Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2709e527-ac99-4042-9e2b-6180dde02c3e · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:39.828957Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4a329ff1-7a7b-42be-8abb-3715d6c5cddc · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 37

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:43.075538Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 47d91345-acc3-4dd4-83ab-563ad1c3d1db · outbound

This paper cites P., Light, M., Shelmerdine, L.

Training language models to be warm and empathetic makes them less reliable and more sycophantic P., Light, M., Shelmerdine, L

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:42.793460Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 7be93626-a516-4606-8595-c8b613f77e11 · outbound

This paper cites How do Large Language Models Navigate Conflicts between Honesty and Helpfulness?.

Training language models to be warm and empathetic makes them less reliable and more sycophantic How do Large Language Models Navigate Conflicts between Honesty and Helpfulness?

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:40.111118Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b5dfc80a-8b63-46cf-8c83-df619672b926 · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 40

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:42.549029Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation b3c09f76-2cb2-47da-855f-4906dc2bb0a7 · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:40.312645Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0ce19684-2723-4e08-9342-675abd7caf10 · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 42

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:42.385459Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 10d4b85f-86cc-4c8d-a6c3-3f2c693e15e7 · outbound

This paper cites ELEPHANT: Measuring and understanding social sycophancy in LLMs.

Training language models to be warm and empathetic makes them less reliable and more sycophantic ELEPHANT: Measuring and understanding social sycophancy in LLMs

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:40.562564Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 583c8a9c-19ba-4e5a-9a58-0d393f33ee11 · outbound

This paper cites A Survey on LLM-as-a-Judge.

Training language models to be warm and empathetic makes them less reliable and more sycophantic A Survey on LLM-as-a-Judge

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-06T12:18:40.635559Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5a7e8985-4fd3-45ee-96fb-2a9f58ee032f · outbound

This paper cites A coefficient of agreement for nominal scales.

Training language models to be warm and empathetic makes them less reliable and more sycophantic A coefficient of agreement for nominal scales

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:42.155723Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 2028af5c-b2eb-40f5-9c6d-041a27e8325b · outbound

This paper cites an unresolved cited work.

Training language models to be warm and empathetic makes them less reliable and more sycophantic Unresolved cited work

Reference 46

Resolution: unresolved; raw_fallback, observed 2026-08-06T12:18:41.900178Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation e334f7f8-f2ae-4228-b687-1d51ba424f89 · outbound

This paper cites et al.Who’s asking? user personas and the mechanics of latent misalignment.

Training language models to be warm and empathetic makes them less reliable and more sycophantic et al.Who’s asking? user personas and the mechanics of latent misalignment

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T12:18:41.698629Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Pith citing papers

Observation 4cfee772-46d1-4f9a-9765-0dd72c65ded6 · inbound

PersonaFuse: A Personality Activation-Driven Framework for Enhancing Human-LLM Interactions cites this paper.

PersonaFuse: A Personality Activation-Driven Framework for Enhancing Human-LLM Interactions Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-04T22:24:43.296659Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 386b8470-7418-4a4d-b447-2cd877437f31 · inbound

Measuring and mitigating overreliance to build human-compatible AI cites this paper.

Measuring and mitigating overreliance to build human-compatible AI Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 63

Resolution: verified exact; arxiv_id, observed 2026-05-21T22:40:43.350299Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 0678f2cd-9d72-460f-b0a2-b67d5f525155 · inbound

From Fact to Judgment: Investigating the Impact of Task Framing on LLM Conviction in Dialogue Systems cites this paper.

From Fact to Judgment: Investigating the Impact of Task Framing on LLM Conviction in Dialogue Systems Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-04T06:48:36.762542Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4752da15-1d6b-4c50-a27f-a65a595337c9 · inbound

Resisting Humanization: Ethical Front-End Design Choices in AI for Sensitive Contexts cites this paper.

Resisting Humanization: Ethical Front-End Design Choices in AI for Sensitive Contexts Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 9

Resolution: verified exact; arxiv_id, observed 2026-05-14T23:59:35.133591Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation bc4b2c06-547f-4f2a-9e45-8174c10dea42 · inbound

Anthropomorphism and Trust in Human-Large Language Model interactions cites this paper.

Anthropomorphism and Trust in Human-Large Language Model interactions Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 32

Resolution: verified exact; arxiv_id, observed 2026-05-15T17:46:23.881202Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 3cb6f21e-c9a6-4474-81dd-0e60ff81509d · inbound

Evaluating the False Trust Engendered by LLM Explanations cites this paper.

Evaluating the False Trust Engendered by LLM Explanations Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 54

Resolution: verified exact; arxiv_id, observed 2026-05-12T07:21:25.726922Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation df56f9c9-1205-47f0-a665-6d78ec0639f4 · inbound

Evaluating the False Trust Engendered by LLM Explanations cites this paper.

Evaluating the False Trust Engendered by LLM Explanations Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 54

Resolution: verified exact; arxiv_id, observed 2026-05-20T21:53:47.132660Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 79f9a07e-49cc-466e-a7ee-b70fb6f97a90 · inbound

What Do People Actually Want From AI? Mapping Preference Plurality cites this paper.

What Do People Actually Want From AI? Mapping Preference Plurality Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 36

Resolution: verified exact; arxiv_id, observed 2026-06-28T01:41:29.719586Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation bc0f54ee-d3be-48d0-b4b6-657ce80eff65 · inbound

A Scalable Approach to Evaluating Moral Sensitivity in LLMs cites this paper.

A Scalable Approach to Evaluating Moral Sensitivity in LLMs Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 108

Resolution: unresolved; no resolver link, observed 2026-07-12T05:44:33.099337Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 58c8329c-dab1-4597-9a46-e2093f08783e · inbound

AI Value Alignment for Evolving Social Norms cites this paper.

AI Value Alignment for Evolving Social Norms Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 80

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:54.881268Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d19d5736-00d0-44a0-a834-a9c36a90c124 · inbound

DelusionEval: Measuring Delusion-Linked Behaviors in AI Chatbots cites this paper.

DelusionEval: Measuring Delusion-Linked Behaviors in AI Chatbots Training language models to be warm and empathetic makes them less reliable and more sycophantic

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-06T11:48:35.902247Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.