Trajectories That Segment Themselves: Agent-Declared Boundaries as a Training Unit

• URL: http://arxiv.org/abs/2608.02302v1
• Date: Mon, 03 Aug 2026 14:27:58 GMT
• Title: Trajectories That Segment Themselves: Agent-Declared Boundaries as a Training Unit
• Abstract summary: We introduce collection-time semantic self-segmentation, in which a declarative contract has the acting agent expose its own boundaries while the trajectory is generated.<n>Because the agent names its conjecture, a reviewer can negate it by name, which lets our protocol manufacture wrong-cause-then-correction transitions that recorded work rarely contains.<n>We then ask what survives deleting the declaration. Given the cut points but not the hypothesis, a model attributes action blocks to their governing hypothesis at over twice chance.<n>Asked instead to place boundaries, a code-blind annotator matches 24 of 40 where random placement matches 11.5, while a mechanical test
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Long-horizon coding-agent trajectories are poorly matched to the credit units available to train on: a single action has no stable value, an episode label merges productive exploration with abandoned directions, and a fixed window cuts where the logging mechanics fall. We introduce collection-time semantic self-segmentation, in which a declarative contract has the acting agent expose its own boundaries while the trajectory is generated. Instantiated with falsifiable causal hypotheses, successive adoptions expose variable-length semantic phases, and no milestone vocabulary, gold patch, environment replay, teacher logits, or retrospective segmenter places a boundary. Because the agent names its conjecture, a reviewer can negate it by name, which lets our protocol manufacture wrong-cause-then-correction transitions that recorded work rarely contains; one collection then yields four supervised targets, including audit supervision from exactly the failed regions an episode label discards. We then ask what survives deleting the declaration. Given the cut points but not the hypothesis, a model attributes action blocks to their governing hypothesis at over twice chance, beating equal-length blocks over the same trajectories (paired sign test $p = 0.0002$), surviving a lexical control and collapsing under label permutation. Asked instead to place boundaries, a code-blind annotator matches 24 of 40 where random placement matches 11.5, while a mechanical test-event rule beats chance at neither end of a strict-to-permissive sweep. The segments are therefore coherent and not cheaply reproducible. Downstream, DPO on 2,551 phase-boundary pairs changes no decision on 91 adversarial held-out items, while four of 60 change on matched-construction items, all wrong to right, where two controls change none: with 1,825 pairs from one generator, the variable to vary next is corpus diversity, not the boundary.

Related papers

• IBIB: A Protocol for Measuring Enterprise AI Systems by Serving Route, Not Model Identifier [0.0]
  All 18 audited benchmarks score advertised model identifiers.<n>A gold-blind capability-binding preflight verifies that a route can execute the evaluation contract before any task reaches it.<n>A reliability-inclusive first-pass scoring rule keeps failure in the score while keeping unsupported capability out.
  arXiv  Detail & Related papers  (2026-09-09T17:31:29Z)
• No Detectable Change in Side-Level WER from Prompt-Level Context: A Preregistered Ablation on a Production Oral-History Corpus [0.0]
  Supplying context at inference time to a large multimodal model is an inexpensive lever for adapting speech transcription to a domain.<n>This work tested that mechanism, in the prompt-conditioning layer of a production oral-history transcription tool, on a sample from its own production corpus.
  arXiv  Detail & Related papers  (2026-08-28T21:26:35Z)
• The Guard That Cried Wolf: How Scary Words Make Agent Guardrails Refuse Legitimate Actions [12.920644711310109]
  Cautious Bench is the first benchmark to make over-safety the construct for agent guardrails.<n>A build-time gate re-derives every example to certify it, so each label is a mechanical consequence of the policy.<n>We find a name-superstition effect: each over-refuses an authorized action more often under a scary-looking object name.
  arXiv  Detail & Related papers  (2026-08-27T11:55:22Z)
• Stage-Replay Divergence Follows the KV Cache: Fixed-Prefix Precision Controls and Bidirectional Cache Transplantation [51.56484100374058]
  Stage-replay reconstructs intermediate token prefixes and treat fresh-prefill continuation as continuation from the decoder state that originally reached the prefix.<n>A matched 200-item experiment compares retained live cache with one-shot prefill of identical integer tokens and places an exact replica on both sides.
  arXiv  Detail & Related papers  (2026-07-30T16:41:40Z)
• What Can Be Enforced? A Theory of Certified Runtime Safety for Tool-Using Agents [0.0]
  guardrails act before irreversible tool calls, but their guarantees depend on what policy state is representable.<n>Pearson representation attacks add a margin, so benign calibration alone does not transfer.<n>Experiments target these distinctions through static diagnostics, controlled-model enumeration, representation rewrites, and paired closed-loop reruns.
  arXiv  Detail & Related papers  (2026-07-24T19:14:26Z)
• Abliteration Is Not a Scalpel: Off-Target Effects of Refusal Removal on Decision Disposition Across Model Families [51.56484100374058]
  Abliteration - deleting a model's refusal direction from its weights - is the standard recipe behind popular "uncensored" open-weight models.<n>As a disposition probe we use 21,600 decisions under uncertainty, replayed through a frozen pipeline so the decision-layer model is the only variable.<n>Three effects replicate across both families (weeks-clustered CIs excluding zero)<n>A fourth effect reverses sign: the same operation makes Gemma-abliterated less confident and Qwen-abliterated more (family CIs non-overlapping)
  arXiv  Detail & Related papers  (2026-07-19T22:27:00Z)
• Legible-by-Construction: Attention and End-to-End Transformers [0.2663471820643486]
  A transformer's feed-forward layer can be rebuilt from explicit fuzzy set operations.<n> hidden units read as named logical operators at no cost to language-model quality.<n>Across five specialized-attention designs at 125M parameters, 44 to 62 percent of value channels become legible.
  arXiv  Detail & Related papers  (2026-07-05T14:08:59Z)
• A Held-Out Transition-Pair Falsifier for Long-Horizon Non-Abelian State Tracking [0.5076419064097734]
  We introduce a held-out transition-pair falsifier for finite non-Abelian group tracking.<n>In a controlled $S_3 times S_3$ benchmark, a projected recurrent state model trained only on length-8 sequences produces error-free final-state predictions.
  arXiv  Detail & Related papers  (2026-06-05T13:26:41Z)
• Survive or Collapse: The Asymmetric Roles of Data Gating and Reward Grounding in Self-Play RL [76.45061154544568]
  Self-play reinforcement learning trains language models on their own generated tasks, co-evolving a proposer and solver without human labels.<n>Recent systems report strong reasoning gains, but collapse and instability are widely observed and poorly understood.<n>We argue instead that self-play stability is governed by two distinct levers: a data-level gate that decides which proposer-generated tasks enter the training pool, and the reward signal that updates the policy on tasks already admitted.
  arXiv  Detail & Related papers  (2026-05-21T09:19:23Z)
• AgentAtlas: Beyond Outcome Leaderboards for LLM Agents [0.025718125188898048]
  AgentAtlas reframes agent evaluation as a diagnostic vocabulary and audit protocol.<n>The paper contributes: (i) a six-state control-decision taxonomy (Act / Ask / Refuse / Stop / Confirm / Recover); (ii) a trajectory-failure vocabulary with primary error source and downstream impact; and (iv) an illustrative protocol study on a synthetic 1,342-item set evaluated with eight models.
  arXiv  Detail & Related papers  (2026-05-19T22:05:12Z)
• To Throw a Stone with Six Birds: On Agents and Agenthood [0.0]
  Six Birds Theory (SBT) treats macroscopic objects as induced closures rather than primitives.<n>We give a type-correct account of agency within SBT.<n>We operationalize this contract in finite controlled systems using four checkable components.
  arXiv  Detail & Related papers  (2026-02-03T10:46:23Z)
• Towards Self-Supervised Covariance Estimation in Deep Heteroscedastic Regression [102.24287051757469]
  We study self-supervised covariance estimation in deep heteroscedastic regression.<n>We derive an upper bound on the 2-Wasserstein distance between normal distributions.<n>Experiments over a wide range of synthetic and real datasets demonstrate that the proposed 2-Wasserstein bound coupled with pseudo label annotations results in a computationally cheaper yet accurate deep heteroscedastic regression.
  arXiv  Detail & Related papers  (2025-02-14T22:37:11Z)
• Ambiguity-Resistant Semi-Supervised Learning for Dense Object Detection [98.66771688028426]
  We propose a Ambiguity-Resistant Semi-supervised Learning (ARSL) for one-stage detectors. Joint-Confidence Estimation (JCE) is proposed to quantifies the classification and localization quality of pseudo labels. ARSL effectively mitigates the ambiguities and achieves state-of-the-art SSOD performance on MS COCO and PASCAL VOC.
  arXiv  Detail & Related papers  (2023-03-27T07:46:58Z)
• Unsupervised Full Constituency Parsing with Neighboring Distribution Divergence [48.69930912510414]
  We propose an unsupervised and training-free labeling procedure by exploiting the property of a recently introduced metric. For implementation, we develop NDD into Dual POS-NDD and build "molds" to detect constituents and their labels in sentences. We show that DP-NDD not only labels constituents precisely but also inducts more accurate unlabeled constituency trees than all previous unsupervised methods with simpler rules.
  arXiv  Detail & Related papers  (2021-10-29T17:27:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.