Recursive Environment Scaling for General Long-Horizon Agents

Lingxiao Jing; Zhengbo Jiao; Yichen Xiang; Bofan Zhu; Lingbei Meng; Yan Ni; Haorui Chen; Ran Ran

cs.AIOct 12, 2026

Recursive Environment Scaling for General Long-Horizon Agents

Authors: Lingxiao Jing, Zhengbo Jiao, Yichen Xiang, Bofan Zhu, Lingbei Meng, Yan Ni, Haorui Chen, Ran Ran,

Organizations: UESTC · HKUST · CUHK (SZ) · THU ·

Abstract

We study environment scaling as a way to increase agent-benchmark difficulty while holding core task objectives fixed. Our framework, Recursive Environment Scaling (RES), transforms existing benchmark environments while preserving their objectives and completion requirements. A proposer changes information access, resource relations, and execution dependencies; execution-grounded checks validate the resulting candidate, and accepted changes are inherited across rounds. This produces related environments for paired evaluation and training-trajectory collection. Experiments on 294 initial tasks from three benchmarks show lower task success for all six evaluation models, with decreases of 57.14--77.14 percentage points on Claw-Eval-Live. With equal trajectory counts and identical optimization settings, fine-tuning on RES trajectories outperforms original-trajectory SFT in all nine model--benchmark comparisons by 7.00--14.61 percentage points. Ablations identify recursive inheritance and dependency composition as important contributors to increased difficulty. Together, these results support environment scaling for both agent evaluation and training.

Figures & tables

Figure 1: Task scaling and recursive environment scaling. Task scaling expands tasks and trajectories. RES preserves core objectives while recursively transforming information access, resource relations, and execution dependencies. Accepted changes accumulate across rounds.

Figure 2: Environment scaling increases execution demand. Historical means for 105 Claw-Eval-Live tasks using DeepSeek-V4.1-Flash (one attempt, 50 turns, success threshold 0.80). Budget

d

averages retained states

0,\ldots,d

, carrying forward early-stopped finals (Appendix C ). Tokens include target input and output; reference nodes count added structure, not shortest solution paths.

Figure 3: Overview of RES. The proposer transforms accepted parent $E_{k}$ . Execution-grounded checks validate the changes, task validity, and dependency feasibility, guiding revisions before accepting a child. Accepted children become parents; target evaluation separately informs stopping.

		Claw-Eval-Live $N=105$		Workspace-Bench-Lite $N=100$		Terminal- Bench 2.1 $N=89$
Model	Environment	Pass@1	Score	TCR@80	RPR	Pass@1
DeepSeek-V4.1-Flash	Original	82.86	87.71	63.00	71.19	91.01
	+RES	5.71	14.20	46.00	60.38	43.82
GPT-5.6 Sol	Original	85.71	91.24	61.00	70.60	88.76
	+RES	26.67	35.44	49.00	65.08	44.94
Claude Opus 5	Original	86.67	92.13	70.00	74.98	89.89

Table 1: Environment scaling lowers task completion across six models and three benchmarks. Original denotes the released benchmark environment; +RES denotes the last accepted environment after scaling. Each model makes one 50-turn attempt per task in both conditions. All metrics use a 0–100 scale. Claw Pass@1 requires native score

\geq 0.80

, and Score is the mean native score. Workspace TCR@80 requires at least 80% of a task’s rubric items to pass; RPR is the percentage of passed rubric items pooled across tasks. Terminal Pass@1 uses binary verifier outcomes.

Claw-Eval-Live ( $N=105$ )
	Qwen3.5-4B			Qwen3.5-9B			Qwen3.5-35B-A3B
Metric	Base	+SFT w/o RES	+SFT w/ RES	Base	+SFT w/o RES	+SFT w/ RES	Base	+SFT w/o RES	+SFT w/ RES
Pass@1	49.52	52.38	61.90	60.00	62.86	71.43	61.90	64.76	77.14
Score	71.51	74.12	79.15	77.75	80.32	81.04	78.96	80.36	82.83
Workspace-Bench-Lite ( $N=100$ )
TCR@80	1.00	4.00	11.00	9.00	12.00	21.00	11.00	14.00	28.00

Table 2: SFT with RES trajectories improves performance on the original benchmarks. For each Qwen3.5 Base model, +SFT w/o RES uses original-environment trajectories; +SFT w/ RES uses only trajectories collected in RES-scaled environments. Both use the same initialization, sample count, and optimization. Each checkpoint makes one 50-turn attempt per task. Metrics use a 0–100 scale: Claw Pass@1 thresholds native scores at 0.80, Score averages native scores, Workspace TCR@80 thresholds task rubric completion at 80%, RPR is the pooled percentage of passed rubric items, and Terminal Pass@1 uses its binary verifier. Bold marks the best result per model.

Variant	Dependency composition	Inheritance	Library	Pass@1 (%) $\downarrow$	Score $\downarrow$
A. Remove dependencies from saved environments (15 tasks)
Saved RES	$\checkmark$	$\checkmark$	Full	0.00	7.73
w/o dependencies	$\times$	$\checkmark$	Full	100.00	92.93
B. Rebuild environments under matched budgets (15 tasks)
Rebuilt RES	$\checkmark$	$\checkmark$	Full	6.67	14.40
w/o inheritance	$\checkmark$	$\times$	Full	93.33	88.73

Table 3: Effects of RES components and recursion depth on Claw-Eval-Live. DeepSeek-V4.1-Flash makes one attempt per environment with a 50-turn limit. Pass@1 uses native score

\geq 0.80

; Score is mean native score on a 0–100 scale. Lower values indicate harder environments. A removes composed dependencies and their prerequisite work from saved environments. B uses the same 15 tasks with matched planned depths and proposer budgets: w/o inheritance evaluates the last independent enhancement from the original environment, and Basic retains six operators. Checkmarks mark enabled components; blue shading identifies full RES controls and the default depth cap of ten. C reevaluates 105 separate lineages, retaining the last accepted environment after early stopping.

Figure 4: Historical performance and interaction demand on Claw-Eval-Live. Lines and bars show 105-seed historical means (Equation 5 ); shading shows task interquartile ranges. The dashed line marks the 50-turn limit. Tokens include target input and output. Termination rates use logged outcomes; actual depth counts appear in Appendix C.2 .

Figure 5: Proposer cost across recursive stages. Bars and lines show the same per-child means on Claw-Eval-Live. Depths 1–4 contain 105, 94, 57, and 18 children; dark bars pool 15 children from five seeds at depths 5–10. Time sums proposer execution durations.

Figure 6: Recursive scaling of a financial reconciliation task. Access and recovery dependencies accumulate while the reconciliation objective is preserved. The diagram shows $E_{0}$ – $E_{2}$ ; endpoint evaluation uses final $E_{3}$ .

Appendix figures & tables12 assets

Supplementary material from the paper’s appendix.

Appendix

	Environment scaling	Depth reevaluation	Model comparison
Attempts per environment	3 at evaluated nodes	1	1
Model-turn limit	60	50	50
Success criterion	Complete task fulfilment	Native score $\geq 0.80$	Native score $\geq 0.80$
Target	DeepSeek-V4.1-Flash	DeepSeek-V4.1-Flash	Six evaluation models
Task cohort	105 seeds	105 seeds / 394 environments	105 matched seeds per model
Primary use	Stopping decisions	Recursive trends	Original vs. RES

Appendix

Table A1: Protocols for environment scaling and independent evaluation on Claw-Eval-Live. Scaling-stage attempts determine when to stop recursion; depth reevaluation measures changes over saved lineages; the paired model comparison evaluates original and RES-scaled environments. A model turn is one target-model response, potentially containing multiple tool calls. Stopping uses complete fulfilment; independent success uses a native-score threshold of 0.80.

Quantity	Mean	Median	P90
Outer proposer executions	1.45	1.00	2.00
Input + output tokens (k)	89.19	65.27	179.82
Summed execution time (s)	112.61	93.56	184.00

Appendix

Table A2: Proposer construction cost per accepted Claw-Eval-Live child. Mean, median, and 90th-percentile (P90) costs cover all 289 children constructed by GPT-5.6 Sol with high reasoning effort. Each child’s costs aggregate its outer executions and retries. Tokens include input and output; time sums execution durations. Target solving and grading are excluded.

Setting	4B-Base	9B-Base	35B-A3B-Base
Peak learning rate	$2\times 10^{-5}$	$10^{-5}$	$5\times 10^{-6}$
Epoch / updates	1 / 527
Maximum sequence length	49,152 tokens
Effective batch size	16 (12 in the last update)
Learning-rate schedule	Cosine, 3.00% warmup (16 steps)
Optimizer	DeepSpeedCPUAdam (AdamW)

Appendix

Table A3: Optimization settings for the Qwen3.5 Base checkpoints in the training comparison. +SFT w/o RES and +SFT w/ RES start independently from the same checkpoint and use these identical settings at each model size. Each condition trains on 8,428 trajectories for one epoch; sequence lengths and effective batch size are measured in tokens and examples, respectively. The checkpoint after the single training epoch is used for evaluation.

Figure A1 : Current retained-state performance and interaction demand. Means and interquartile ranges cover all 105 seeds, retaining early-stopped finals without averaging earlier budgets.

Figure A2 : Current versus historical benchmark means. Historical means average the retained measurements from budgets zero through the current budget.

Figure A3 : Historical interaction demand at selected budgets. Per-task means over retained states

0,\ldots,d

are averaged equally across 105 seeds, including failed and turn-limited trajectories. Tokens include repeated input context.

$d$	Reach $d$	Continue	Pass@1	Score	Turns	Tools	Tokens (M)
0	105	105	87/105	87.84	3.55	8.05	0.01
1	105	94	63/105	66.69	33.71	151.79	1.66
2	94	57	18/105	25.17	46.91	217.47	2.85
3	57	18	8/105	15.30	48.81	236.37	3.00
4	18	5	6/105	13.65	49.00	235.34	3.02
5	5	4	4/105	11.84	49.10	236.54	3.04

Appendix

Table A4: Performance and interaction cost across recursion-depth budgets on Claw-Eval-Live. DeepSeek-V4.1-Flash receives one 50-turn attempt per environment. Each budget

d

covers 105 tasks, retaining stopped environments. Reach

d

counts chains attaining depth

d

; Continue counts chains exceeding it. Pass@1 counts scores

\geq 0.80

; Score is the native mean on a 0–100 scale. Turns, Tools, and Tokens are task means; tokens sum target input and output in millions.

Figure A4 : Performance across recursion-depth caps. Pass@1 (orange) and mean native score (blue) use a 0–100 scale over 105 seeds, retaining stopped environments. Table 3 selects caps 0, 1, 2, 3, 5, and 10.

Task	$d_{i}$	Original	Saved	w/o deps.	Rebuilt	w/o rec.	Basic
A04	10	93.00	0.00	89.00	0.00	97.00	95.00
D02	4	96.00	0.00	96.00	98.00	99.00	98.00
COMM_13	5	100.00	4.00	82.00	3.00	100.00	82.00
DATA_09	8	86.00	7.00	93.00	3.00	88.00	88.00
PRODAPP_01	3	92.00	14.00	99.00	15.00	99.00	97.00
DATA_07	3	96.00	0.00	96.00	3.00	100.00	98.00

Appendix

Table A5: Per-task native scores for the 15-task component ablations. DeepSeek-V4.1-Flash receives one 50-turn attempt per environment; scores use a 0–100 scale, with success at 80 or above. Original is the unmodified task. Saved and Rebuilt denote full RES controls; w/o deps. removes dependencies from Saved, w/o rec. independently enhances Original and evaluates the predeclared last construction, and Basic rebuilds with six operators. The rebuilt variants share planned depth

d_{i}

Figure A5 : Per-task effects of the component interventions. Filled blue markers show the corresponding full RES control; hollow orange markers show the ablation. Segments connect the same task, and dashed lines mark the 80-point success threshold. All panels use the task order in Table A5 .

Condition	Turns	Tool calls	Tokens (k)	Relative tokens ( $\times$ )
Original	2.87	6.67	8.43	1.00
Saved RES	50.00	226.93	3115.42	369.40
w/o dependencies	19.07	85.60	723.76	85.82
Rebuilt RES	49.73	163.93	1644.20	194.96
w/o inheritance	21.20	90.40	651.09	77.20
Basic library	18.53	39.60	227.01	26.92

Appendix

Table A6: Target execution cost for the 15-task component comparisons. Values are task means for DeepSeek-V4.1-Flash with one 50-turn attempt per environment, including failed and turn-limited attempts. Saved RES is the dependency-removal control; Rebuilt RES is shared by the independent-construction and basic-library comparisons. Tokens sum target input and output; k denotes thousands. Relative tokens divides the condition’s mean by the Original mean before rounding. Proposer construction and grading costs are excluded.

Condition	$\tau=0.70$	$\tau=0.80$	$\tau=0.90$
Original	15/15	15/15	13/15
Saved RES	0/15	0/15	0/15
w/o dependencies	15/15	15/15	10/15
Rebuilt RES	1/15	1/15	1/15
w/o inheritance	14/15	14/15	12/15
Basic library	14/15	12/15	10/15

Appendix

Table A7 : Success counts under alternative native-score thresholds on Claw-Eval-Live. Each entry counts tasks with score

\geq\tau

over the cohort. Component rows use 15 paired tasks; depth rows use 105 reevaluated seeds with stopped environments retained. Counts use saved scores.

AI Agent Benchmarks

AI agent benchmarks measure how reliably autonomous systems plan, use tools, interact, and complete tasks across controlled and real-world settings.

See all 1,503 papers

Scalable Environments Drive Generalizable Agents
May 18, 2026Jiayi Zhang, Fanqi Kong, Guibin Zhang +7
HKUST(GZ) · DeepWisdom · PKU +3
AgentCE-Bench: Agent Configurable Evaluation with Scalable Horizons and Controllable Difficulty under Lightweight Environments
Apr 7, 2026Wang Yang, Chaoda Song, Xinpeng Li +7AI Agent Benchmarks Agent Benchmarks
Case Western Reserve University · NII LLMC (Japan) · University of Zurich
AutoEnvScaling: Automating the Data Flywheel with Terminal Agents
Oct 9, 2026Simon Yu, Baolin Peng, Bo Liu +8LLM Agent Training Self-Evolving Agents
Northeastern University · Microsoft · Stanford University +2

May 18, 2026J/K move · Enter open · S save

Scalable Environments Drive Generalizable Agents

Jiayi Zhang, Fanqi Kong, Guibin Zhang, Maojia Song+6

HKUST(GZ) · DeepWisdom · PKU+3

Open paper AskPDF arXiv

Generalizable agents should adapt to diverse tasks and unseen environments beyond their training distribution. This position paper argues that such generalization requires environment scaling: expanding the distribution of executable rule-sets that agents interact with, rather than only increasing trajectories or tasks within fixed benchmarks. Current scaling practices largely focus on collecting more experience or broader task sets under fixed interaction rules, leaving agents brittle when underlying interfaces, dynamics, observations, or feedback signals change. The core challenge is therefore a world-level distribution shift: agents need systematic exposure to environments with meaningfully different executable rule-sets. To clarify this challenge, we propose a unified taxonomy that separates trajectory scaling, task scaling, and environment scaling by their primary deliverables and by what changes in the executable rule-set. Building on this taxonomy, we synthesize construction paradigms for scalable environments, contrasting programmatic generators that prioritize controllability and verifiability with generative world models that offer broader coverage and open-endedness. We further outline how environment scaling can be coupled with stateful learning mechanisms, emphasizing learned update rules for cross-environment adaptation. We conclude by discussing alternative perspectives and argue that scalable environments provide the essential substrate for measurable and controllable progress toward robust general agents.

Abstract

AI Agent Benchmarks

Explore similar work

Scalable Environments Drive Generalizable Agents

AgentCE-Bench: Agent Configurable Evaluation with Scalable Horizons and Controllable Difficulty under Lightweight Environments

AutoEnvScaling: Automating the Data Flywheel with Terminal Agents