ForceBench: Benchmarking Force Integration for Robotic Manipulation

Xuan Han; Zinuo Jin; Tianlin Li; Yunfan Lou; Xin Liu; Yuchen Xue; Yuanpeng He; Boxian Yao

cs.ROOct 14, 2026

ForceBench: Benchmarking Force Integration for Robotic Manipulation

Authors: Xuan Han, Zinuo Jin, Tianlin Li, Yunfan Lou, Xin Liu, Yuchen Xue, Yuanpeng He, Boxian Yao,

Organizations: PKU-PI Lab, Peking University · National University of Singapore · University of California, Berkeley · ePyBot Intelligence

Abstract

Force feedback provides vision-language-action (VLA) policies with physical observations that can guide action selection and execution. However, differences in tasks, demonstrations, and control settings across existing systems hinder reliable comparisons of force integration designs. We introduce ForceBench, a simulation benchmark for evaluating six-axis wrist-force integration under a unified protocol. ForceBench combines eight tasks spanning physical-property inference, resistance-dependent interaction, and contact-rich assembly with 1,600 synchronized demonstrations and task-specific success criteria. We compare one no-force baseline and six force integration designs on a shared $π_{0.5}$ backbone. Prompt Fusion achieves the highest observed mean simulation success of 35.31%, compared with 23.25% for the baseline, and improves success rates across all three task categories. Predictive and temporal gains in simulation concentrate on physical-property inference, while both temporal designs underperform the baseline on assembly. Real-world evaluation of eight systems across five tasks supports the main qualitative trends: Prompt Fusion and Dual-Path Fusion achieve 63.00% mean success, compared with 40.00% for the baseline, but only Prompt Fusion improves every task. These findings support ForceBench's practical value for identifying the benefits and limitations of force integration across manipulation demands. The project page is available at https://forcebench-platform.github.io/.

Figures & tables

Fig. 2: Simulation tasks. (a) Weight Sort and (b) CoM Sort require physical-property inference; (c) Nut Thread requires contact-rich assembly; (d) Faucet Turn and Faucet Direction, (e) Stove Turn and Stove Press, and (f) Drawer Pull require resistance-dependent interaction. Insets mark the hidden property or resistance transition that appearance does not reveal; Faucet Direction and Stove Press share the faucet and stove environments, respectively.

Fig. 3: Force integration on a shared

\pi_{0.5}

backbone. 1 – 4 denote Cross-Attention, Token Fusion + Pred, Prompt Fusion, and Token Fusion, respectively;

\pi_{0.5}

(No Force) denotes the baseline. 5 and 6 show the additional Temporal Prompt and Dual-Path Fusion references. Alternative pathways are compared separately. The future-force head in 2 provides training-only supervision. Force and torque components together form the six-axis force observation.

Method	Weight Sort	CoM Sort	Faucet Turn	Faucet Direction	Stove Turn	Stove Press	Drawer Pull	Nut Thread	Mean $\uparrow$
$\pi_{0.5}$ (No Force) [ 3 ]	54.0	45.5	9.5	31.5	8.5	1.0	4.5	31.5	23.25
$\pi_{0.5}$ ( $+$ Token Fusion	50.5	50.0	12.0	28.5	6.0	2.5	13.0	26.0	23.56
$\pi_{0.5}$ ( $+$ Prompt Fusion	88.0	56.5	21.0	41.5	16.0	2.5	23.0	34.0	35.31
$\pi_{0.5}$ ( $+$ Cross-Attention	54.5	51.0	6.5	33.0	5.5	2.5	6.5	17.5	22.13
$\pi_{0.5}$ ( $+$ Token Fusion + Pred	78.0	48.5	13.5	32.5	6.0	1.0	4.0	26.5	26.25
$\pi_{0.5}$ ( $+$ Temporal Prompt	90.0	69.5	17.5	31.0	7.0	21.5	2.5	22.0	32.63

TABLE I: Simulation task success rates (%), with 200 episodes per task–method pair. Mean averages the eight tasks equally.

Fig. 4: Task outcomes and state diagnostics (200 episodes per task–method pair). Bars show success (green) and labeled failure categories (ochre/purple). In (a,b,d), filled/open markers report success rates for the labeled physical conditions. In (c,e,f,g,h), dots represent episode measurements colored by success (green) or failure (ochre). In (c,e,g), shaded target intervals map to

[0,1]

; reaching these intervals alone does not establish full success. In (f), angles are normalized by the direction-specific resistance-peak angle. In (h), offset is measured during descent at a nut height of 35 mm on a symmetric-log axis; the ellipsis omits tick labels only. Count pairs report full-task successes / episodes reaching the target assembly state (centered, upright, and at the target height); full success additionally requires release and withdrawal.

Method	Weight Sort	Plate Retrieve	Stamp Press	Plug Insert	Block Insert	Mean $\uparrow$
	Property inference	Resistance-dependent interaction			Assembly
$\pi_{0.5}$ (No Force)	30.0	20.0	40.0	60.0	50.0	40.00
Token Fusion	45.0	40.0	50.0	60.0	10.0	41.00
Prompt Fusion	70.0	40.0	65.0	80.0	60.0	63.00
Cross-Attention	60.0	30.0	40.0	50.0	55.0	47.00
Token Fusion + Pred	45.0	60.0	60.0	45.0	35.0	49.00

TABLE II: Real-world task success rates (%), with 20 trials per task–method pair. The mean weights all five tasks equally; bold denotes the highest rate in each column.

Panel of Figure 6 — Fig. 5: Real-world tasks. (a) Plug Insert, (b) Block Insert, (c) Plate Retrieve, (d) Stamp Press, and (e) Weight Sort. Each panel shows key interaction frames from the external camera.

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Method	Property	Resistance	Assembly	Total
No Force	49.75	11.00	31.50	372
Token Fusion	50.25	12.40	26.00	377
Prompt Fusion	72.25	20.80	34.00	565
Cross-Attention	52.75	10.80	17.50	354
Token Fusion + Pred	63.25	11.40	26.50	420
Temporal Prompt	79.75	15.90	22.00	522

Appendix

TABLE III: Simulation category means (%) and total successes, derived from Table I . Each total is out of 1,600 episodes. Property, resistance, and assembly contain two, five, and one task, respectively.

Fig. 7: Six-axis wrench trajectories from three successful simulation demonstrations, each selected at seed 0 (episode 1). Top:

F_{x},F_{y},F_{z}

in newtons. Bottom:

T_{x},T_{y},T_{z}

in newton-metres. Gray intervals mark the logged lift-and-weigh phase in (a), resistance-rise phase in (b), and thread-engagement phase in (c). All 457, 530, and 917 control-rate samples are shown, respectively. These are illustrative expert trajectories, not comparisons between learned policies.

Fig. 8: Six-axis wrench trajectories covering all five real-world tasks. Each row shows one recorded demonstration: (a) Weight Sort (full can), (b) Plate Retrieve, (c) Stamp Press, (d) Plug Insert, and (e) Block Insert. Left:

F_{x},F_{y},F_{z}

in newtons. Right:

T_{x},T_{y},T_{z}

in newton-metres. All 424, 376, 675, 442, and 684 samples are retained, respectively, including offsets and transient variations. Time is relative to the first recorded sample; vertical scales vary by task and quantity. Colors and line styles follow Fig. 7 . The source records are episodes 0, 0, 1, 0, and 0 of the corresponding archived task collections, and are illustrative demonstrations rather than policy-evaluation rollouts.

Robotic Manipulation

Robotic manipulation studies how robots perceive, plan, and control physical interactions to grasp, move, assemble, and use objects.

See all 1,685 papers

IMBench: A Benchmark for Intuitive Robotic Manipulation
Jul 17, 2026Anurag Maurya, Sukhvansh Jain, Prajwal Avhad +9Contact-Rich Robotic Manipulation Robotic Manipulation
Manav Robotics, Bengaluru, India
ForceVLA2: Unleashing Hybrid Force-Position Control with Force Awareness for Contact-Rich Manipulation
Mar 16, 2026Yang Li, Zhaxizhuoma, Hongru Jiang +11Contact-Rich Robotic Manipulation Force-Controlled Robotic Manipulation
Tongji University · Shanghai AI Laboratory · Lumos Robotics +6
ForceBand: Learning Forceful Manipulation with sEMG
Jun 24, 2026Botao He, Zhi Wang, Linna Kuang +8Robot Skill Learning Contact-Rich Robotic Manipulation
Amazon FAR · University of Maryland · Johns Hopkins University

Jul 17, 2026J/K move · Enter open · S save

IMBench: A Benchmark for Intuitive Robotic Manipulation

Anurag Maurya, Sukhvansh Jain, Prajwal Avhad, Gautham Balachandran+8

Manav Robotics, Bengaluru, India

Contact-Rich Robotic Manipulation Robotic Manipulation Robot Manipulation Benchmarks VLMs for Robotics

Open paper AskPDF arXiv

Humans combine reasoning and motor control to solve complex manipulation tasks under diverse constraints. They build an understanding of the physical world that helps them convert reasoning into actions and quickly adapt to new scenes, tasks, and rules. We refer to this capability as intuitive manipulation. Existing benchmarks fail to capture this integration: they evaluate physical reasoning in isolation from execution, or measure policy performance without requiring explicit reasoning. We introduce IMBENCH, a benchmark designed to evaluate intuitive manipulation as an integrated capability spanning perception, physical reasoning, action generation, and iterative execution. Our tasks require models to infer task-relevant physical structure and generate feasible action sequences under explicit constraints, including contact-rich manipulation, tool use, and multi-stage dependencies. We introduce a benchmark of 35 tasks, 14K filtered trajectories, and scalable tools for generating diverse scenarios. Experiments reveal a consistent gap: vision language models show partial physical reasoning ability but fail to produce executable plans, while state-of-the-art vision-language-action models struggle to satisfy task constraints and generalize across scenarios. These results identify intuitive manipulation as a missing axis in current foundation models and generalist robot policies, and position IMBENCH as a step toward evaluating and enabling more integrated, adaptive physical intelligence.

Abstract

Robotic Manipulation

Explore similar work

IMBench: A Benchmark for Intuitive Robotic Manipulation

ForceVLA2: Unleashing Hybrid Force-Position Control with Force Awareness for Contact-Rich Manipulation

ForceBand: Learning Forceful Manipulation with sEMG