Pith. sign in

Paper Citation Record · LEDGER

We Can't Understand AI Using our Existing Vocabulary

As of 9 August 2026, this Paper Citation Record lists 57 of 57 outbound references and 7 inbound Pith citation observations for arXiv:2502.07586.

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

pith.paper-citation-record.v1
2502.07586 v1

Coverage vector

measured 57 of 57 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-08T12:17:46.099027Z

measured 64 of 64 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-09T06:31:02.800959+00:00

measured 7 of 7 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-07T10:20:07.722478Z

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

57 of 57 outbound references displayed

  • verified exact1
  • verified fuzzy26
  • unresolved30
  • parse uncertain0
  • malformed identifier0
  • metadata mismatch0

External citation measurements

1
arxiv_reference, observed 2026-08-05T02:28:24.338817Z

Outbound references

Observation f0114f40-f29e-493c-8e4f-f2a3ffe7f77a · outbound

This paper cites Sanity checks for saliency maps.

We Can't Understand AI Using our Existing Vocabulary Sanity checks for saliency maps

Reference 1

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.641191Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.910826Z digest=sha256:2d5a2fabc7a1f713c9bf8ae1e6fb73570f48e5d2fb3c3f06bbe5017b7d3f4fb4

Observation ba94dd62-519a-4efa-aaf8-e9e7a1472e39 · outbound

This paper cites Understanding intermediate layers using linear classifier probes.

We Can't Understand AI Using our Existing Vocabulary Understanding intermediate layers using linear classifier probes

Reference 2

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.914983Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.914983Z digest=sha256:53ec7f11b81292dbbf36894dee61f599b10478112db35a3a7a0c6ca340bde795

Observation e345c04f-97e8-4850-9efb-e04579c099af · outbound

This paper cites Soft prompting might be a bug, not a feature, 2023.

We Can't Understand AI Using our Existing Vocabulary Soft prompting might be a bug, not a feature, 2023

Reference 3

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.631707Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.918814Z digest=sha256:b1e42bfc76932ed0469c4feb19538f97d35374f4e77fc2ac948ba2066523a098

Observation c9012803-10d5-4e02-88a0-827077157ca9 · outbound

This paper cites Network dissection: Quantifying interpretability of deep visual representations.

We Can't Understand AI Using our Existing Vocabulary Network dissection: Quantifying interpretability of deep visual representations

Reference 4

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.622387Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.922813Z digest=sha256:c77106bdd61e6fe38355a7c44a534eb95dbfa144c7939fcec7360b2da47da86e

Observation b8448aba-237d-43e4-b627-ae5a546ad272 · outbound

This paper cites W., and Kim, B.

We Can't Understand AI Using our Existing Vocabulary W., and Kim, B

Reference 5

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.612362Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.926343Z digest=sha256:6312e1b54dbd0352e6e3d5c404390802569d26733301b5f6a0db075c11c474c3

Observation 6972cf4a-e87e-4339-a21b-b353e1b510a7 · outbound

This paper cites an unresolved cited work.

We Can't Understand AI Using our Existing Vocabulary Unresolved cited work

Reference 6

Resolution
unresolved
raw_fallback, observed 2026-08-08T12:17:46.604309Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.929817Z digest=sha256:cfec476a29e1917f1ee54abc8927f25678c80e2ae428721e4ea9673a11f88cc4

Observation 5133694b-32a6-4655-9724-0916febd92b5 · outbound

This paper cites Evaluating the Susceptibility of Pre-Trained Language Models via Handcrafted Adversarial Examples.

We Can't Understand AI Using our Existing Vocabulary Evaluating the Susceptibility of Pre-Trained Language Models via Handcrafted Adversarial Examples

Reference 7

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.934346Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.934346Z digest=sha256:e66cd28228f023f9cc457b92adea1bcbb9cdb9000e437fd5a315fc63dc65b2da

Observation f85ec66c-fba3-4f55-acf6-05a6dc36f14d · outbound

This paper cites Large Language Monkeys: Scaling Inference Compute with Repeated Sampling.

We Can't Understand AI Using our Existing Vocabulary Large Language Monkeys: Scaling Inference Compute with Repeated Sampling

Reference 8

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.937878Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.937878Z digest=sha256:59bc7159dd1bcb1a2de5405b887afbb82990091b0eadc6af7264960c248f0664

Observation 2204f1e1-5c7b-4f78-a8ec-44bded843c7f · outbound

This paper cites The comparative psychology of artificial intelligences, May 2019.

We Can't Understand AI Using our Existing Vocabulary The comparative psychology of artificial intelligences, May 2019

Reference 9

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.594022Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.940775Z digest=sha256:3a64a2e130ce8bfd9aad17544846e60fff9b494295149885a5e0b5ddfdc86045

Observation 4ed0d734-d9c5-4c39-9ebe-ad22d8630653 · outbound

This paper cites Discovering latent knowledge in language models without supervision.

We Can't Understand AI Using our Existing Vocabulary Discovering latent knowledge in language models without supervision

Reference 10

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.582860Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.943609Z digest=sha256:b88a43be1c2c09c58260b20877254dc7ac74c3b5189cc5859ba5a164b4013abe

Observation e9fecbd3-35ae-4914-9d9b-74103094f7ae · outbound

This paper cites J., Jarrell, T.

We Can't Understand AI Using our Existing Vocabulary J., Jarrell, T

Reference 11

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.571720Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.947423Z digest=sha256:f729e588862d5c46419803c7b118a5126fff1d56cc16eba7b4720cfcad6295cf

Observation ef45a9ee-3dda-4212-9301-c51ea85fedaf · outbound

This paper cites Sparse Autoencoders Find Highly Interpretable Features in Language Models.

We Can't Understand AI Using our Existing Vocabulary Sparse Autoencoders Find Highly Interpretable Features in Language Models

Reference 12

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.950284Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.950284Z digest=sha256:e83b63805f101308d04f7552f4a62b27b20923f93762d8b1923cae10ba7ff2d8

Observation ddab1732-8eb1-4dac-a74e-241eab9a29c6 · outbound

This paper cites Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment.

We Can't Understand AI Using our Existing Vocabulary Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment

Reference 13

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.953953Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.953953Z digest=sha256:cdfe92d1bd3197a92e5d4ad06194d06bbcc558018bb6f25940de0d7ca322b3df

Observation 38108302-dbe4-4f64-828b-a887e68f3529 · outbound

This paper cites Towards A Rigorous Science of Interpretable Machine Learning.

We Can't Understand AI Using our Existing Vocabulary Towards A Rigorous Science of Interpretable Machine Learning

Reference 14

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.957466Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.957466Z digest=sha256:aa570617885e607f12c8248e2c7608ba131cc9d13e650972039ac11ed21f9c76

Observation df64c77f-51d4-41ad-8cb3-9ea25294a122 · outbound

This paper cites Probing for incremental parse states in autoregressive language models.

We Can't Understand AI Using our Existing Vocabulary Probing for incremental parse states in autoregressive language models

Reference 15

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.961327Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.961327Z digest=sha256:ed92774751d19d35045449144a9c9cbd7363c0cab87786ac2be9eade901cc298

Observation 5ed264c6-b9cf-4b53-85fd-de99f045ab79 · outbound

This paper cites Probing for semantic evidence of composition by means of simple classification tasks.

We Can't Understand AI Using our Existing Vocabulary Probing for semantic evidence of composition by means of simple classification tasks

Reference 16

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.964554Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.964554Z digest=sha256:75068f9b150198170d811b22744a6351a139e59a6ed8783747877280ff91d7c9

Observation 4659becd-9631-4dac-a800-6e579ba8777c · outbound

This paper cites Craft: Concept recursive activation factorization for explainability.

We Can't Understand AI Using our Existing Vocabulary Craft: Concept recursive activation factorization for explainability

Reference 17

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.967743Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.967743Z digest=sha256:de102ac668aa894268ec02d8f816ff8ab15e5e2913d1429cb0b65edefc931186

Observation aa9da2c4-78b7-477f-941f-710dcea49cfa · outbound

This paper cites Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context.

We Can't Understand AI Using our Existing Vocabulary Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

Reference 18

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.970269Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.970269Z digest=sha256:d9fd65232013d5f6f7cf5b1db28a1d34fd79799aa4c919bcb97b2ae4d1588638

Observation 78260703-aedf-4ca4-b5ac-ff3232fb7645 · outbound

This paper cites Interpretation of neural networks is fragile.

We Can't Understand AI Using our Existing Vocabulary Interpretation of neural networks is fragile

Reference 19

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.553728Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.973514Z digest=sha256:d3b82fed5b11adce9527d4cdb118aae29c71cd8b8da532e240293c03bdc8a29d

Observation 76219a77-dd93-40ca-be8f-225222e77e70 · outbound

This paper cites Y., and Kim, B.

We Can't Understand AI Using our Existing Vocabulary Y., and Kim, B

Reference 20

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.976892Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.976892Z digest=sha256:1a509d9d54400ba74a63b4b7b5530f795a58d11d43983354eeb5b72d4eabb055

Observation ffc424f4-be32-465e-906b-1ed193474941 · outbound

This paper cites and Liang, P.

We Can't Understand AI Using our Existing Vocabulary and Liang, P

Reference 21

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.980012Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.980012Z digest=sha256:2d83ba302f095a738445ff0acf420a469e76cf884d0ec0c24be0f4ee4d802554

Observation 7d150d0a-87c5-4a07-8446-8ddd55d422b6 · outbound

This paper cites and Manning, C.

We Can't Understand AI Using our Existing Vocabulary and Manning, C

Reference 22

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.537955Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.983441Z digest=sha256:b1100c6933e3501d39c0d2864b96810719b15d34c252590af35c4a46bf06eabc

Observation c649df2e-ccc2-4157-a84a-2a3cc67d0357 · outbound

This paper cites J., yelong shen, Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., and Chen, W.

We Can't Understand AI Using our Existing Vocabulary J., yelong shen, Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., and Chen, W

Reference 23

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.986577Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.986577Z digest=sha256:8e0b0f9d057b0befdc308d879909d3548cb6fd841510a4fbdee3711d4ee8a28e

Observation 9ce079bc-3280-4acb-8d97-2acb22e5976e · outbound

This paper cites Beyond interpretability: developing a language to shape our relationships with AI , Apr 2022.

We Can't Understand AI Using our Existing Vocabulary Beyond interpretability: developing a language to shape our relationships with AI , Apr 2022

Reference 24

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.520470Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.989917Z digest=sha256:adfa0e67cc4a80a1d9ee55bdb1b9306e8db774509646f1cfd57fa7ff4b60b235

Observation 8d0b08e1-bb8c-402e-8719-bab1b24674f5 · outbound

This paper cites u tt , K. T., D \.

We Can't Understand AI Using our Existing Vocabulary u tt , K. T., D \

Reference 25

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.509938Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.992922Z digest=sha256:3733aa4921adeb7142ce9e53b951ea8bd5ce509f6a782467b397bbc8f811f382

Observation 2cd44cb7-20fd-4804-aee0-3f9ddba56348 · outbound

This paper cites The emergence of number and syntax units in LSTM language models.

We Can't Understand AI Using our Existing Vocabulary The emergence of number and syntax units in LSTM language models

Reference 26

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:45.995991Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:45.995991Z digest=sha256:1f8d6130e71a223b51256fcb940dc18abe528670286716e1816972c18ec5563b

Observation 94fefb0b-8e34-4a5a-af9e-90920866616c · outbound

This paper cites T., Isola, P., Globerson, A., Irani, M., and Mosseri, I.

We Can't Understand AI Using our Existing Vocabulary T., Isola, P., Globerson, A., Irani, M., and Mosseri, I

Reference 27

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.497297Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:45.999231Z digest=sha256:4369f3c05a303ff8571d2af578ba8a08607bfbd5ca101917cb478d305d927921

Observation db58d6ff-1d89-4d2b-8333-a4d1b4841704 · outbound

This paper cites The Power of Scale for Parameter-Efficient Prompt Tuning.

We Can't Understand AI Using our Existing Vocabulary The Power of Scale for Parameter-Efficient Prompt Tuning

Reference 28

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.002888Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.002888Z digest=sha256:40d9b2e88e7bc9d0b4806a5d88b47e086aecf8250da14a94d0cbd2b731dcd249

Observation 04bab83c-b424-42ae-9dae-0023354899de · outbound

This paper cites The Mythos of Model Interpretability.

We Can't Understand AI Using our Existing Vocabulary The Mythos of Model Interpretability

Reference 29

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.006051Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.006051Z digest=sha256:daeb62ccc1d776988d6f7f7e7cac64cf3d2162e93492ecb6e5ebae5d5e7b807f

Observation 77bfd67b-f294-4019-9399-d45a40498cc5 · outbound

This paper cites Evaluation beyond task performance: analyzing concepts in alphazero in hex.

We Can't Understand AI Using our Existing Vocabulary Evaluation beyond task performance: analyzing concepts in alphazero in hex

Reference 30

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.486118Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.008760Z digest=sha256:a24992c76d208628fa23456c7ece4c298410e1c9ccf6f6ad100675eda26d151c

Observation 8a2c813b-1a02-4ef4-9155-a463a3d5be24 · outbound

This paper cites an unresolved cited work.

We Can't Understand AI Using our Existing Vocabulary Unresolved cited work

Reference 31

Resolution
unresolved
raw_fallback, observed 2026-08-08T12:17:46.474962Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.011954Z digest=sha256:8af594bbc54d87edf8efe6ddcbadfc7b4e16f419671ad84de98a626f2a755baa

Observation 575709f8-4997-4653-bf74-b49dce32a580 · outbound

This paper cites and Tegmark, M.

We Can't Understand AI Using our Existing Vocabulary and Tegmark, M

Reference 32

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.015958Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.015958Z digest=sha256:7409ce6a52f4dbc78ced2668eacbac3fe0e4fd355f8729fa022be2bf55cb351d

Observation 425b88b1-8213-499b-a31d-a0cdf4be7c7d · outbound

This paper cites Acquisition of chess knowledge in alphazero.

We Can't Understand AI Using our Existing Vocabulary Acquisition of chess knowledge in alphazero

Reference 33

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.454629Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.019166Z digest=sha256:8380471353c4c810e1ddcd10288aa54f19dc437dc21a55bc5d63d5c20c2c439c

Observation eb18a884-77e9-4fd3-8010-4e1bb4897c2a · outbound

This paper cites Gemma: Open Models Based on Gemini Research and Technology.

We Can't Understand AI Using our Existing Vocabulary Gemma: Open Models Based on Gemini Research and Technology

Reference 34

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.022713Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.022713Z digest=sha256:b0c6c7a9e729a56c21e6930a27e1019c9ef6ec64ec8b2b63e245fd74ceeeee5f

Observation 6ebd649a-d77d-4008-946c-4fda04fbd4a5 · outbound

This paper cites Mechanistic interpretability, variables, and the importance of interpretable bases.

We Can't Understand AI Using our Existing Vocabulary Mechanistic interpretability, variables, and the importance of interpretable bases

Reference 35

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.443681Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.026817Z digest=sha256:0d4b04b4dacf4bef47ae559d350165a4fc5037bef32cf5e7a5e2eae6bb2e6535

Observation afd70683-1744-4394-9703-ce71776e2bfa · outbound

This paper cites Unsupervised sentiment neuron.

We Can't Understand AI Using our Existing Vocabulary Unsupervised sentiment neuron

Reference 36

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.431940Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.029725Z digest=sha256:15a31dcec067a49c8e737762669e9876b4a4556bcc01ea02ebac3529e85afdbd

Observation 79ef4637-969f-41d8-8e1e-60cafda7dac8 · outbound

This paper cites Training language models to follow instructions with human feedback.

We Can't Understand AI Using our Existing Vocabulary Training language models to follow instructions with human feedback

Reference 37

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.033350Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.033350Z digest=sha256:78b3f19aecb9a3d3f8dae51556d98e5c2dced60aea6ce365b9c3c2f070148866

Observation 8a9d19fd-5d2a-47ee-9698-bef35d984bd7 · outbound

This paper cites D., Ermon, S., and Finn, C.

We Can't Understand AI Using our Existing Vocabulary D., Ermon, S., and Finn, C

Reference 38

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.036419Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.036419Z digest=sha256:9c02fa74548472a22941a314028b22bbd2fbad5abb038713215e7196a5554d42

Observation ae4dd5fa-601d-4700-9f76-db59e157d0c8 · outbound

This paper cites Concept Alignment as a Prerequisite for Value Alignment.

We Can't Understand AI Using our Existing Vocabulary Concept Alignment as a Prerequisite for Value Alignment

Reference 39

Resolution
verified exact
local_arxiv, observed 2026-08-08T12:17:46.190080Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.039441Z digest=sha256:ca55a5ad93af839c848c0ff34c82977e985e33634e4daea089a8304d405c9175

Observation eb9f8d8f-892a-49c7-b7b8-1c3e8d6e5f55 · outbound

This paper cites Bridging the Human-AI Knowledge Gap: Concept Discovery and Transfer in AlphaZero.

We Can't Understand AI Using our Existing Vocabulary Bridging the Human-AI Knowledge Gap: Concept Discovery and Transfer in AlphaZero

Reference 40

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.043419Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.043419Z digest=sha256:cd9799681a2c6af268f3738470a8d13066c898c5a83a1e84b58cf3fb1e746145

Observation 04d36f49-eb66-47d2-b5e1-8d12d8d6d4bb · outbound

This paper cites R., Cogswell , M., Das , A., Vedantam , R., Parikh , D., and Batra , D.

We Can't Understand AI Using our Existing Vocabulary R., Cogswell , M., Das , A., Vedantam , R., Parikh , D., and Batra , D

Reference 41

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.408563Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.046803Z digest=sha256:71a915eb6b533884214ed0eec0bcbba7161ac7363b8038bba3d302a0681a4637

Observation ba8c12df-dae5-4f36-915d-3a93056fee6f · outbound

This paper cites and Stern, M.

We Can't Understand AI Using our Existing Vocabulary and Stern, M

Reference 42

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.399206Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.049695Z digest=sha256:9c5a9bd9067a69fad4448b3fc015e83f1b7ccd369e2a21e6b6842c7338a12b7d

Observation 56f8760b-0b74-4be8-8555-33cf5af26cf3 · outbound

This paper cites Does string-based neural mt learn source syntax? In Conference on Empirical Methods in Natural Language Processing, 2016.

We Can't Understand AI Using our Existing Vocabulary Does string-based neural mt learn source syntax? In Conference on Empirical Methods in Natural Language Processing, 2016

Reference 43

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.389328Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.052578Z digest=sha256:96259f0e22e2a492ff7aafb4cd9544f9e641d68c03402f41f6ac9dca1a3b1cc3

Observation 9ea5e4c2-36b6-4a86-93db-8092f974bb7b · outbound

This paper cites Learning important features through propogating activation functions.

We Can't Understand AI Using our Existing Vocabulary Learning important features through propogating activation functions

Reference 44

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.379641Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.055438Z digest=sha256:4fe16ae0ed78fea836284dc5994b46094df2e763e53ba6cf2829f48deaf7b911

Observation 058bfed7-c6c2-4d82-ac31-7778c2ce6739 · outbound

This paper cites Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm.

We Can't Understand AI Using our Existing Vocabulary Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm

Reference 45

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.058042Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.058042Z digest=sha256:80cd6565f7929704eab8e92cdf0a76ffe4426d76caaa1cab3333d223c619812d

Observation 15105908-7ae7-4c0b-a9a0-729423558b5c · outbound

This paper cites SmoothGrad : Removing noise by adding noise.

We Can't Understand AI Using our Existing Vocabulary SmoothGrad : Removing noise by adding noise

Reference 46

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.370361Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.061140Z digest=sha256:bfb67f8277ba0d42cf5fec657d60542ecd211269b243d2971ea3c6b6c6310c9b

Observation fab15114-8883-401d-bb4c-47caca59770a · outbound

This paper cites D., Ng, A., and Potts, C.

We Can't Understand AI Using our Existing Vocabulary D., Ng, A., and Potts, C

Reference 47

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.064148Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.064148Z digest=sha256:92a0931058f6ded78a00bed0de8eee5c5d21739635cc0aadcdc0d1a54c754ff4

Observation 48bc5a67-2aad-446d-9143-d795ea84ef0d · outbound

This paper cites Axiomatic attribution for deep networks.

We Can't Understand AI Using our Existing Vocabulary Axiomatic attribution for deep networks

Reference 48

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.355473Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.067263Z digest=sha256:0e62cc1f6d9fdd6cfbde5b08184f844db98f8044852b6d0c6b08f09bb049522a

Observation 95f5a988-5d14-4016-9720-efa559dc8689 · outbound

This paper cites The bitter lesson.

We Can't Understand AI Using our Existing Vocabulary The bitter lesson

Reference 49

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.071018Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.071018Z digest=sha256:87092c5744c9a6ac1ecb4f39d9e53e813531c5933f067e347ded91dc620c8364

Observation b4387d1f-b9e1-45a1-a69a-c5676ab9f5c6 · outbound

This paper cites T., Kim, N., Durme, B.

We Can't Understand AI Using our Existing Vocabulary T., Kim, N., Durme, B

Reference 50

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.340592Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.073882Z digest=sha256:6e97cf1f3fd90a8d6a059bcf6184ba568edbf6cd12792c3df642e0c01972c140

Observation 7f6339a6-7742-4207-86d7-90a1e0dfc20e · outbound

This paper cites Sanity checks for saliency metrics.

We Can't Understand AI Using our Existing Vocabulary Sanity checks for saliency metrics

Reference 51

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.331177Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.077168Z digest=sha256:5cedcb1ae40d2c05b26911f071fe5942b225b89917392ee81f2d0078fe147c05

Observation bf2eae36-013c-4fdb-9f47-1fa914e2373a · outbound

This paper cites an unresolved cited work.

We Can't Understand AI Using our Existing Vocabulary Unresolved cited work

Reference 52

Resolution
unresolved
raw_fallback, observed 2026-08-08T12:17:46.321513Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.080856Z digest=sha256:d0e2568f1d6bf8a644aaf56f3f48055faaa37bdb2de4f132e85bcf00963b8393

Observation c73f075a-526d-40e1-8728-f510f8f93d4e · outbound

This paper cites In two moves A lphago and L ee S edol redefined future.

We Can't Understand AI Using our Existing Vocabulary In two moves A lphago and L ee S edol redefined future

Reference 53

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.311380Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.084643Z digest=sha256:bc0398537da10caf8ded7aecdf7330574e1f9dcfb84cc65aba272c77dfe52743

Observation 7fe23c39-cc43-4b14-9aa3-a6423c0d64eb · outbound

This paper cites Tractatus Logico-Philosophicus.

We Can't Understand AI Using our Existing Vocabulary Tractatus Logico-Philosophicus

Reference 54

Resolution
verified fuzzy
raw_fallback, observed 2026-08-08T12:17:46.301094Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-08-08T12:17:46.087798Z digest=sha256:78e4cfe4c1f399499cecfc1977fb98a6325c5e450809e2ed29830c8397ef35b8

Observation fb57445c-104e-4d29-8366-a41b0b1394ba · outbound

This paper cites LIMA: Less Is More for Alignment.

We Can't Understand AI Using our Existing Vocabulary LIMA: Less Is More for Alignment

Reference 55

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.091708Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.091708Z digest=sha256:f3903af5981450fe54ef8a45bca38d5f9b97d62a6373636f6856d1b8f57ca215

Observation c891e88c-8a2c-41da-b346-cf95f653c2e0 · outbound

This paper cites Representation Engineering: A Top-Down Approach to AI Transparency.

We Can't Understand AI Using our Existing Vocabulary Representation Engineering: A Top-Down Approach to AI Transparency

Reference 56

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.094968Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.094968Z digest=sha256:bb8e0726aa6a922647844cdefb741a79dfccdae8d7cdcade49ab040670f6d3b6

Observation 9bca0746-ae3e-411c-87a8-eeb390a90136 · outbound

This paper cites write newline.

We Can't Understand AI Using our Existing Vocabulary write newline

Reference 57

Resolution
unresolved
no resolver link, observed 2026-08-08T12:17:46.099027Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-08T12:17:46.099027Z digest=sha256:05bd1dac719940ab7365fabb9a6298688ce7141724e3531587188043de24eb59

Pith citing papers

Observation 223baed2-020a-48be-a7d5-d5b89ddb9d3a · inbound

When Models Know More Than They Can Explain: Quantifying Knowledge Transfer in Human-AI Collaboration cites this paper.

When Models Know More Than They Can Explain: Quantifying Knowledge Transfer in Human-AI Collaboration We Can't Understand AI Using our Existing Vocabulary

Reference 19

Resolution
unresolved
no resolver link, observed 2026-08-07T10:20:07.722478Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T10:20:07.722478Z digest=sha256:7c6f9f69008075a4533f863482008f20bc18ee0c0fe4a6ba894032ca59fd9d51

Observation 2bd2d3f5-d96d-47ae-85b0-ee97be0b6d7f · inbound

Because we have LLMs, we Can and Should Pursue Agentic Interpretability cites this paper.

Because we have LLMs, we Can and Should Pursue Agentic Interpretability We Can't Understand AI Using our Existing Vocabulary

Reference 2025

Resolution
unresolved
no resolver link, observed 2026-08-07T01:03:21.013201Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T01:03:21.013201Z digest=sha256:4fcebf4f46a04375fa1703ab8638a8fe9f9261aad6dc129f84cc30cf155f22f1

Observation 2944d7c0-1570-4ee7-b869-a250d97d33af · inbound

Prompting as Scientific Inquiry cites this paper.

Prompting as Scientific Inquiry We Can't Understand AI Using our Existing Vocabulary

Reference 12

Resolution
unresolved
no resolver link, observed 2026-08-06T21:25:08.505585Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-06T21:25:08.505585Z digest=sha256:8cff58c2fe7102c5ab9a3b909addcc4575e888ae2404d35b4cb65224c96b02e2

Observation a43955ff-fd20-4b02-b378-2682f2cbce73 · inbound

CHiQPM: Calibrated Hierarchical Interpretable Image Classification cites this paper.

CHiQPM: Calibrated Hierarchical Interpretable Image Classification We Can't Understand AI Using our Existing Vocabulary

Reference 19

Resolution
verified exact
arxiv_id, observed 2026-05-17T04:29:02.016115Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-05-17T04:24:33.101394Z digest=sha256:d47afc7467d8015278ec50d41ca48733b0534e0f5bfcb0b340e393363416f635

Observation 824ce3ba-a3db-40c0-ac14-3d76a12fec4e · inbound

LLMs Should Express Uncertainty Explicitly cites this paper.

LLMs Should Express Uncertainty Explicitly We Can't Understand AI Using our Existing Vocabulary

Reference 2

Resolution
verified exact
arxiv_id, observed 2026-05-10T22:10:48.070207Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-05-10T20:12:46.367760Z digest=sha256:5ec5cbe629a4e79ecdec41fef7d65fe415fe26316745ed53bad2e001756fa76c

Observation ad89b66f-8078-44f5-95b9-04e27d425bc6 · inbound

LLMs Should Express Uncertainty Explicitly cites this paper.

LLMs Should Express Uncertainty Explicitly We Can't Understand AI Using our Existing Vocabulary

Reference 2

Resolution
verified exact
arxiv_id, observed 2026-05-15T07:15:11.883765Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-05-15T07:11:39.471878Z digest=sha256:dfcd20f7640c29a8a4c033438cb85de32f88600246c48f854a6596e6e6ba1b33

Observation 6861df02-5cef-43dc-9b18-8e9b5dbf732b · inbound

ToxiREX: A Dataset on Toxic REasoning in ConteXt cites this paper.

ToxiREX: A Dataset on Toxic REasoning in ConteXt We Can't Understand AI Using our Existing Vocabulary

Reference 230

Resolution
verified exact
arxiv_id, observed 2026-06-29T04:43:06.959640Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-06-29T04:33:18.794505Z digest=sha256:46118f8f2ddd10662ac68bd75f8a50de92e2769b846483a9ca2da1ddefac46aa