Paper Citation Record · LEDGER

Learning Safety Constraints for Large Language Models

As of 9 August 2026, this Paper Citation Record lists 25 of 25 outbound references and 5 inbound Pith citation observations for arXiv:2505.24445.

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

pith.paper-citation-record.v1
2505.24445 v1

Coverage vector

measured 25 of 25 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T12:30:16.628557Z

measured 30 of 30 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 5 of 5 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-02T11:57:51.928730Z

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

25 of 25 outbound references displayed

verified exact1
verified fuzzy2
unresolved20
parse uncertain0
malformed identifier2
metadata mismatch0

External citation measurements

0
arxiv_reference, observed 2026-08-05T02:28:24.338817Z

Outbound references

Observation 211c39db-1a9e-44e8-952e-e3a33be7cf9e · outbound

This paper cites MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialogues.

Learning Safety Constraints for Large Language Models MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialogues

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:14.985484Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 70a87e9a-0d49-4527-b423-512a812a02e6 · outbound

This paper cites SequenceMatch: Imitation Learning for Autoregressive Sequence Modelling with Backtracking.

Learning Safety Constraints for Large Language Models SequenceMatch: Imitation Learning for Autoregressive Sequence Modelling with Backtracking

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:15.354687Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7363c4c8-40c6-482c-b8e0-7d7bc2524288 · outbound

This paper cites Safe Exploration in Continuous Action Spaces.

Learning Safety Constraints for Large Language Models Safe Exploration in Continuous Action Spaces

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:15.456333Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 08849323-40da-4714-a7fe-c4502c5a2c33 · outbound

This paper cites RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models.

Learning Safety Constraints for Large Language Models RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:15.620283Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 519a09b1-99c1-4f8d-bf1e-1b428aa01b35 · outbound

This paper cites Measuring Massive Multitask Language Understanding.

Learning Safety Constraints for Large Language Models Measuring Massive Multitask Language Understanding

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:15.758569Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d78c2c74-df8f-49c9-87d0-4dd07e0a2639 · outbound

This paper cites Backdoor Attacks for In-Context Learning with Language Models.

Learning Safety Constraints for Large Language Models Backdoor Attacks for In-Context Learning with Language Models

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:15.821630Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 56cec1a0-9d3c-46d4-b90a-bb3b4cf916c4 · outbound

This paper cites AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models.

Learning Safety Constraints for Large Language Models AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:15.939144Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f1d5073-9f8a-45ce-be38-8a0bd2e8880e · outbound

This paper cites Enhancing LLM Safety via Constrained Direct Preference Optimization.

Learning Safety Constraints for Large Language Models Enhancing LLM Safety via Constrained Direct Preference Optimization

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:16.012101Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 41797f56-e347-427e-879c-d3aa95e037af · outbound

This paper cites Mpax: Mathematical pro- gramming in jax.

Learning Safety Constraints for Large Language Models Mpax: Mathematical pro- gramming in jax

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:16.069748Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5df8c693-4082-40b4-82ae-d1d98aff73cb · outbound

This paper cites Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization.

Learning Safety Constraints for Large Language Models Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:16.143593Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2481352e-529d-4429-8213-38d98983fd1c · outbound

This paper cites Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming.

Learning Safety Constraints for Large Language Models Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:16.217942Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b459100a-c042-471c-be37-2187899fe70f · outbound

This paper cites Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs.

Learning Safety Constraints for Large Language Models Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:16.287165Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 74b81ec1-c15b-4fb3-a0bb-fb078d10d9bd · outbound

This paper cites Token-level Direct Preference Optimization.

Learning Safety Constraints for Large Language Models Token-level Direct Preference Optimization

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:16.341645Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d0b25abb-0b16-4501-8622-7fd70047f051 · outbound

This paper cites Panacea: Pareto Alignment via Preference Adaptation for LLMs.

Learning Safety Constraints for Large Language Models Panacea: Pareto Alignment via Preference Adaptation for LLMs

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:16.406578Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cc3e2173-fffc-4d37-aa3e-989ffb38cc47 · outbound

This paper cites Beyond one-preference-fits-all alignment: Multi-objective direct preference optimization.

Learning Safety Constraints for Large Language Models Beyond one-preference-fits-all alignment: Multi-objective direct preference optimization

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T12:30:17.936203Z

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.

Observation c06c74f7-cc17-4d49-8621-04f06d5d164c · outbound

This paper cites an unresolved cited work.

Learning Safety Constraints for Large Language Models Unresolved cited work

Reference 23

Resolution: unresolved; raw_fallback, observed 2026-08-07T12:30:17.725886Z

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.

Observation 947a66c1-85a2-4efe-b390-8d096005b368 · outbound

This paper cites {human question}\n{model answer}.

Learning Safety Constraints for Large Language Models {human question}\n{model answer}

Reference 24

Resolution: malformed identifier; raw_fallback, observed 2026-08-07T12:30:17.579256Z

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.

Observation e2f0e159-4e83-4861-9165-53506c0bd2b9 · outbound

This paper cites Results show mean ± standard deviation.

Learning Safety Constraints for Large Language Models Results show mean ± standard deviation

Reference 25

Resolution: malformed identifier; raw_fallback, observed 2026-08-07T12:30:17.301765Z

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.

Observation 12ed9f93-7ac7-444e-845a-2394f397f838 · outbound

This paper cites Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks.

Learning Safety Constraints for Large Language Models Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks

Reference 1999

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:14.786704Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3fc8dcb9-694d-40f7-bfa6-435116fa0bca · outbound

This paper cites TruthfulQA: Measuring How Models Mimic Human Falsehoods.

Learning Safety Constraints for Large Language Models TruthfulQA: Measuring How Models Mimic Human Falsehoods

Reference 2014

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:15.888021Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5691e74e-0f86-4b94-8d91-7297b889fcc0 · outbound

This paper cites Interpreting Neural Networks through the Polytope Lens.

Learning Safety Constraints for Large Language Models Interpreting Neural Networks through the Polytope Lens

Reference 2016

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:15.138252Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 62406f3f-8d92-4d13-9b40-ac547bef0427 · outbound

This paper cites AI Control: Improving Safety Despite Intentional Subversion.

Learning Safety Constraints for Large Language Models AI Control: Improving Safety Despite Intentional Subversion

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:15.698528Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4c4f088e-3a56-4d72-9850-0fc309a42dff · outbound

This paper cites Unlocking Decoding-time Controllability: Gradient-Free Multi-Objective Alignment with Contrastive Prompts.

Learning Safety Constraints for Large Language Models Unlocking Decoding-time Controllability: Gradient-Free Multi-Objective Alignment with Contrastive Prompts

Reference 2022

Resolution: verified exact; local_arxiv, observed 2026-08-07T12:30:17.032993Z

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.

Observation 28a20280-8c32-43e8-b0aa-224937fb62cc · outbound

This paper cites Defending Against Unforeseen Failure Modes with Latent Adversarial Training.

Learning Safety Constraints for Large Language Models Defending Against Unforeseen Failure Modes with Latent Adversarial Training

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-07T12:30:15.271464Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05e6391c-18c6-4105-a685-75cfa16f0747 · outbound

This paper cites and Bartlett, P.

Learning Safety Constraints for Large Language Models and Bartlett, P

Reference 2024

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T12:30:18.150799Z

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.

Pith citing papers

Observation e2bd960a-5dbb-4c95-9548-c457ff5dc172 · inbound

When control meets large language models: From words to dynamics cites this paper.

When control meets large language models: From words to dynamics Learning Safety Constraints for Large Language Models

Reference 257

Resolution: verified exact; arxiv_id, observed 2026-05-21T14:54:13.126908Z

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.

Observation 9036905c-b097-4870-b700-ba09c4020dc2 · inbound

Preference Instability in Reward Models: Detection and Mitigation via Sparse Autoencoders cites this paper.

Preference Instability in Reward Models: Detection and Mitigation via Sparse Autoencoders Learning Safety Constraints for Large Language Models

Reference 7

Resolution: verified exact; arxiv_id, observed 2026-05-20T22:49:10.216730Z

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.

Observation c9ba4192-0e7a-4ee5-ab2c-d234dde2bced · inbound

Defending Jailbreak Attacks on Large Language Models via Manifold Trajectory Kinetics cites this paper.

Defending Jailbreak Attacks on Large Language Models via Manifold Trajectory Kinetics Learning Safety Constraints for Large Language Models

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-06-27T22:01:20.847532Z

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.

Observation 511e8842-49c9-4e3a-a11b-475e12eb343f · inbound

Greedy Coordinate Diffusion: Effective and Semantically Coherent Adversarial Attacks via Diffusion Guidance cites this paper.

Greedy Coordinate Diffusion: Effective and Semantically Coherent Adversarial Attacks via Diffusion Guidance Learning Safety Constraints for Large Language Models

Reference 15

Resolution: metadata mismatch; arxiv_id, observed 2026-07-03T17:08:43.438995Z

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.

Observation 9d400a3f-0fe4-42db-8206-794618102f72 · inbound

Geometry-Guided Constraint Learning for LLM Safety Classification cites this paper.

Geometry-Guided Constraint Learning for LLM Safety Classification Learning Safety Constraints for Large Language Models

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-02T11:57:51.928730Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.