ATC: Agent Trajectory Control for Role-Contract Enforcement in Black-Box Multi-Agent Systems
Organizations: University of Oklahoma · University of Florida
Abstract
Multi-agent LLM systems assign different responsibilities to different agents: one plans, another retrieves evidence, another verifies, another executes. This division of labor is only meaningful if each agent stays within the functions it is authorized to perform. Over many turns, however, an agent can produce a locally reasonable message that exercises authority belonging to another role: a verifier that begins planning, a retriever that asserts unsupported conclusions, an executor that changes state no one approved. We formalize this as role drift, a turn-level breach of an agent's role contract. We introduce Agent Trajectory Control (ATC), which makes each role's authority explicit as a contract and enforces it at the point where a message would enter the shared conversation. ATC scores a proposed output for out-of-role deviation, leakage into the functions of other roles, impermissible speech acts, and logical contract breaches, then computes a leakage-regularized target that constrains regeneration; the revised message is re-checked before admission. Enforcement acts on the message rather than the generator, so ATC governs agents whose internal representations are unavailable. Across homogeneous and heterogeneous teams, ATC improves task performance and role consistency while lowering role-drift energy relative to strong baselines spanning prompting, memory, workflow control, and activation editing, with the largest gains under explicit role-confusion stress tests and in larger, more diverse teams.
Figures & tables
| Metric | Stable | Drifted |
|---|---|---|
| 0.18 | 0.57 | |
| 0.11 | 0.49 | |
| 0.07 | 0.36 | |
| 0.05 | 0.31 | |
| 0.36 | 1.47 | |
| out–leak corr. | 0.76 | |
| Setting | Method | EntCollabBench | MultiAgentBench | SWE-bench Verified | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Primary | RC | RDE | Primary | RC | RDE | Primary | RC | RDE | ||
| S1: Hom-10 | Single-Agent | 54.8 | 68.1 | 0.39 | 49.6 | 65.4 | 0.42 | 27.4 | 63.2 | 0.45 |
| Prompt-Only | 61.3 | 72.6 | 0.34 | 55.2 | 70.8 | 0.36 | 31.8 | 68.7 | 0.39 | |
| Struct-Role | 64.7 | 78.9 | 0.28 | 58.4 | 76.1 | 0.31 | 33.9 | 74.5 | 0.33 | |
| A-MEM | 66.5 | 77.3 | 0.27 | 60.1 | 75.6 | 0.30 | 35.6 | 73.8 | 0.32 | |
| CAA | 65.2 | 79.8 | 0.26 | 59.3 | 78.4 | 0.29 | 34.8 | 76.2 | 0.31 | |
| Variant | Primary | RC | RDE |
|---|---|---|---|
| Full ATC | 76.4 | 93.5 | 0.11 |
| w/o role subspace term | 68.2 | 80.1 | 0.27 |
| w/o leakage penalty | 70.4 | 84.3 | 0.22 |
| w/o linguistic term | 72.1 | 87.5 | 0.19 |
| w/o logical contract | 71.6 | 86.8 | 0.20 |
| w/o regeneration | 69.7 | 82.9 | 0.24 |
| Correction | Primary | RC | RDE |
|---|---|---|---|
| No correction | 59.2 | 67.4 | 0.41 |
| Hard role reminder | 62.9 | 72.6 | 0.35 |
| Pure projection | 71.3 | 86.1 | 0.20 |
| Hard orthogonality to | 69.8 | 87.0 | 0.18 |
| Full ATC objective | 76.4 | 93.5 | 0.11 |
| Method | Role- confuse | Authority- swap | Late- correction |
|---|---|---|---|
| Prompt-Only | 42.5 | 39.8 | 44.1 |
| Struct-Role | 49.3 | 46.7 | 50.2 |
| A-MEM | 51.1 | 48.5 | 55.8 |
| SEA | 54.7 | 51.6 | 56.3 |
| SOP-Role | 57.9 | 55.2 | 58.6 |
| ATC | 70.8 | 68.4 | 72.1 |
| Primary | RC | RDE | |
|---|---|---|---|
| 0.00 | 71.3 | 86.1 | 0.20 |
| 0.25 | 74.2 | 90.4 | 0.15 |
| 0.50 | 76.4 | 93.5 | 0.11 |
| 1.00 | 75.8 | 92.7 | 0.12 |
| 2.00 | 73.6 | 90.8 | 0.15 |
| Setting | Primary | RC | RDE |
|---|---|---|---|
| Hom-5 | 72.9 | 92.1 | 0.12 |
| Hom-10 | 74.6 | 91.8 | 0.13 |
| Het-10 | 75.8 | 92.7 | 0.12 |
| Het-12 | 76.4 | 93.5 | 0.11 |
| Het-16 | 75.1 | 91.4 | 0.14 |
| Method | Calls | Tok. | Lat. | RCG/C |
|---|---|---|---|---|
| Prompt-Only | 12.4 | 116.0 | 24.3 | 0.00 |
| Struct-Role | 12.6 | 100.2 | 24.8 | 0.52 |
| A-MEM | 15.8 | 137.5 | 31.2 | 0.34 |
| CAA | 12.8 | 119.8 | 27.1 | 0.62 |
| SEA | 13.1 | 121.4 | 29.3 | 0.76 |
| SOP-Role | 13.6 | 126.2 | 28.5 | 0.86 |
Appendix figures & tables10 assets
Supplementary material from the paper’s appendix.
Appendix
| Setting | Agents | Families | Role Assignment |
|---|---|---|---|
| S1: Hom-10 | 10 | 1 | same backbone for all roles |
| S2: Het-10 | 10 | 5 | role groups by model family |
| S3: Het-12 | 12 | 5+ | S2 + auditor, specialist |
| Parameter | Value |
|---|---|
| Temperature | 0.2 |
| Top- | 0.95 |
| Max output tokens / turn | 512 |
| Max turns / episode | 200 |
| Context budget | 128K tokens |
| Retrieval budget | top- |
| Symbol | Value | Meaning |
|---|---|---|
| 1.00 | out-of-role deviation weight | |
| 1.00 | cross-role leakage weight | |
| 0.50 | linguistic violation weight | |
| 0.75 | logical violation weight | |
| 0.50 | leakage penalty in correction | |
| 0.95 | temporal memory factor |
| Agent | Drifted Output | ATC Correction | Why Corrected |
|---|---|---|---|
| Verifier | “We should assign the retrieval task to the Memory-Reader, then ask the Executor to submit the final answer.” | “The proposed plan requires verification. I will check whether the retrieval step matches the task constraints and whether the Executor has sufficient evidence before finalization.” | Verifier leaked into Planner role. ATC suppresses planning behavior and restores verification behavior. |
| Retriever | “The contract likely requires approval from the finance lead, so the answer is yes.” | “I found the relevant policy passage but cannot infer the final decision. The Verifier should check whether the approval condition is satisfied.” | Retriever generated an unsupported conclusion instead of returning evidence. |
| Executor | “I will update the shared memory to mark this request as approved.” | “I can produce the final response only after the Memory-Writer updates the state and the Verifier confirms the approval condition.” | Executor attempted an unauthorized state update. ATC enforces the logical role contract. |
| Memory-Writer | “The user should choose option B because it is cheaper.” | “I will store the updated constraint: option B has the lowest cost among the available choices.” | Memory-Writer began making a recommendation. ATC restricts it to state maintenance. |
| Behavior encoder | Primary | RC | RDE | Relative encoding latency |
|---|---|---|---|---|
| BAAI/bge-large-en-v1.5 | 76.4 | 93.5 | 0.11 | 1.00 |
| BAAI/bge-base-en-v1.5 | 75.2 | 91.8 | 0.14 | 0.62 |
| BAAI/bge-small-en-v1.5 | 73.6 | 89.4 | 0.17 | 0.34 |
| Correction policy | Corrected turns (%) | Primary | RC | RDE | Tokens (K) | Latency (s) |
|---|---|---|---|---|---|---|
| SOP-Role (no runtime correction) | 0.0 | 66.8 | 78.2 | 0.29 | 126.2 | 28.5 |
| ATC-Gated, | 24.7 | 71.3 | 85.1 | 0.21 | 128.1 | 29.7 |
| ATC-Gated, | 46.8 | 73.9 | 89.2 | 0.16 | 129.9 | 30.8 |
| ATC-Gated, | 72.6 | 75.6 | 92.1 | 0.13 | 131.5 | 31.7 |
| ATC (correction at every turn) | 100.0 | 76.4 | 93.5 | 0.11 | 132.8 | 32.4 |
| Leakage weight | Primary | RC | RDE | Harmful over-correction | Content preserved |
|---|---|---|---|---|---|
| 76.4 | 93.5 | 0.11 | 5.6% | 94.3% | |
| 73.6 | 90.8 | 0.15 | 18.4% | 82.1% |
LLM Agents
Large language model agents use language models to plan, use tools, and act across multi-step tasks and environments.
See all 757 papers