Literature
Display
← All papers

ATOM-Bench: A Real-World Benchmark for Atomic Skills and Compositional Generalization in Manipulation Policies

Zenan Wu · Bingqing Wei · Lu Liu · Zheqi He · Xi Wang · Jiakang Liu · Zehui Li · Guocai Yao · Jing-Shu Zheng · Xi Yang · Yongtao Wang

Preprint posted · CoRL 2026 (accepted) ↗

Robotics9 min text · CondensedOriginal paper ↗
Contents

Overview

Project page. This edition follows the complete twenty-one-page, 15 June 2026 v1. All fifty-four illustrated task definitions, eight figures, twelve numbered tables and both diagnostic equations are retained. Author spelling and order follow the PDF title page.

At a glance

ATOM-Bench fine-tunes manipulation policies on atomic tasks, then tests whether they can reuse those skills in unseen task combinations on physical robots. Each of two platforms has fifteen atomic and twelve compositional tasks; 3,000 human demonstrations support adaptation and five policies undergo 2,700 main evaluation rollouts. Simple instruction grounding is generally stronger than precise motor execution, and compositional completion remains low even for the strongest adapted policy. The proposed scores diagnose gaps, but do not by themselves identify the causal fraction of failures due to composition.

1 Introduction

Successful task demonstrations and broad benchmark averages do not reveal whether a policy acquired reusable skills. ATOM-Bench controls which atomic tasks appear during fine-tuning and withholds their tested combinations. This supplies a more explicit adaptation/evaluation split than a general task-suite ranking.

The contribution is a paired physical benchmark, not a new control policy. Its controlled holdout concerns this fine-tuning set; the study does not establish that similar skills or combinations were absent from every foundation model’s pretraining data.

Complete original PDF block S1.F1
Figure 1: Overview of ATOM-Bench. Tabletop manipulation tasks are factorized into motor atoms and instruction atoms. Policies are fine-tuned on atomic tasks and evaluated on both atomic skill acquisition and held-out compositional generalization.

2 Related Work

Simulation suites such as VIMA, LIBERO and VLABench enable controlled variation; LIBERO-PRO and RoboHiMan probe generalization. Physical benchmarks including RoboArena, AutoEval, RoboChallenge and GM-100 emphasize ranking, scale or task diversity.

ATOM-Bench adds an explicit atomic-adaptation/composition-evaluation protocol. Its policy comparison includes π₀.₅, LingBot-VLA, GR00T N1.6, SmolVLA and Motus, under the authors’ fine-tuning recipes rather than a training-free test of base checkpoints.

3 ATOM-Bench

3.1 Task Factorization and Task Groups

Six motor families describe physical operations: pick-and-place, reorientation, pushing, stacking, pouring and articulated access. Seven instruction atoms describe color, shape, size, source relation, destination, counting and exclusion. Each platform instantiates these in eight motor tasks, seven instruction tasks and twelve withheld compositions.

Instruction atoms use a simple executable carrier action rather than a language-only test. The paired dual-arm tasks add roles, coordination and ordered stages. These are practical functional categories, not a proof that every task varies only one independent factor; for example, an access sequence can contain placing an object inside a drawer.

3.2 Data Collection

The single-arm setup uses a seven-joint Franka Panda, Robotiq 2F-85 gripper and left/front/wrist RealSense views. Cobot Magic uses two arms with left/right/front views and master-arm teleoperation, following the Mobile ALOHA hardware lineage. Franka demonstrations use a SpaceMouse.

One hundred demonstrations per atomic task yield 1,500 per platform and 3,000 total, recorded at 30 Hz. Collection takes about nine hours for Franka and 12.5 hours for Cobot Magic. Composition demonstrations are excluded from fine-tuning.

Complete original PDF block S3.F2
Figure 2: Data collection platforms used in ATOM-Bench. Left: the Franka Panda with single-arm. Right: the Agilex Cobot Magic with dual-arm.

3.3 Metrics

SR counts complete task success. PSR averages the fraction of annotated intermediate score points achieved, so it can be high when a policy repeatedly completes only early steps. Each task uses ten shared physical test seeds and reported aggregates are task macro-averages.

AS is the mean isolated-task PSR over the atoms required by a composition, averaging representative tasks when an atom has several. CFS normalizes the positive AS-minus-composition-PSR gap by remaining process failure. AS is an empirical baseline, not a proven ceiling on composed success; CFS is an algebraic diagnostic, not an identified causal share. Sequential attrition, context changes and different score-point structures can create a gap without isolating a new composition mechanism. The PSR=1 case needs a denominator convention that is not specified.

CFS is computed per task before averaging, so applying the nonlinear formula to table-level mean AS/PSR need not reproduce the reported mean CFS. Transfer Gap is single-task SR minus joint-atomic SR: positive favors specialization and negative favors joint training.

Equations
AS⁡(X)=1|A⁡(X)|​∑a∈A⁡(X)PSR⁡(a).\mathrm{AS}(X)=\frac{1}{|A(X)|}\sum_{a\in A(X)}\mathrm{PSR}(a).
CFS⁡(X)=max⁡(0,AS⁡(X)−PSR⁡(X))1−PSR⁡(X).\mathrm{CFS}(X)=\frac{\max(0,\mathrm{AS}(X)-\mathrm{PSR}(X))}{1-\mathrm{PSR}(X)}.

4 Experiments

The experiments separate atomic acquisition, withheld composition, observed failure types and the effect of joint versus single-task adaptation. The main total equals five policies × two platforms × twenty-seven tasks × ten trials = 2,700 rollouts; the specialization and robustness studies are additional protocols.

4.1 Experimental Setup

For each platform, each policy is jointly fine-tuned on all fifteen atomic tasks. Its checkpoint is then tested on those tasks and the twelve withheld compositions. The single-task comparison trains a separate checkpoint on one hundred demonstrations of each task and is reported only for π₀.₅.

A mask overlay helps operators reproduce the same ten starting arrangements across policies, but placement is approximate rather than metrologically exact. These are physical scene seeds, not evidence of ten independent training runs. Model-specific optimization budgets and adaptation modes remain part of the comparison.

4.2 Atomic Skill Acquisition

π₀.₅ leads Table 1: instruction SR is 94.3% on Franka and 71.4% on Cobot Magic, versus motor SR 46.2% and 45.0%. Motus is generally second; LingBot-VLA slightly exceeds Motus on Franka motor tasks but falls behind on the dual-arm track. PSR often exceeds SR, exposing incomplete execution.

The qualitative pattern is clear—pick-and-place and simple attributes are easier than fine control, counting and exclusion—but the per-atom means differ between text and Figure 3. For pick_place/pour/push/reorient/access, the prose gives 74/20/16/11/9%, while the chart labels show 69/16/14/10/8%. Both are preserved rather than silently reconciled. Instruction tasks also use simpler carrier actions, so their stronger results do not isolate reasoning from motor difficulty.

Complete original PDF block S4.T1
Table 1: Atomic skill acquisition under atomic-transfer fine-tuning. Mean SR and PSR (%) over the Motor and Instruction Sets.
Original selectable table
Franka Panda Cobot Magic
Motor Instruction Motor Instruction
Model SR PSR SR PSR SR PSR SR PSR
Pi0.5 46.2 56.2 94.3 95.7 45.0 72.0 71.4 83.2
Motus 36.2 46.2 67.1 78.7 35.0 59.3 50.0 67.5
LingBot-VLA 37.5 46.5 54.3 60.5 26.2 42.8 21.4 47.1
GROOT N1.6 28.8 47.1 57.1 69.5 23.8 41.0 27.1 52.1
SmolVLA 17.5 29.8 11.4 31.2 10.0 32.2 5.7 29.3
Complete original PDF block S4.F3
Figure 3: Per-atom Success Rate under atomic-transfer fine-tuning. Each bar is one model’s SR over all tasks instantiating that atom on both platforms.

4.3 Compositional Generalization

π₀.₅ has AS 83.3%/79.5% but compositional SR only 15.8%/16.7% on Franka/Cobot Magic; its compositional PSR is 30.4%/42.6%. Motus achieves SR 10.8%/7.5%, and the remaining policies are lower. Thus isolated partial-progress competence does not translate directly into reliable completion of these new combinations.

The reported CFS values are 73.7%/56.8% for π₀.₅ and about 49%/48% for Motus. These summarize normalized process gaps under the proposed metric, rather than proving that those fractions of rollouts failed solely because of composition. Comparison of AS with SR also mixes partial-progress and full-completion measures; Table 2 retains both PSR and SR for that reason.

Complete original PDF block S4.T2
Table 2: Atomic-to-compositional transfer. Mean SR, PSR, AS, CFS (%) over the Composition Set, per platform. AS is the atomic-baseline ceiling; CFS is the share of compositional failure attributable to composition rather than to weak atoms.
Original selectable table
Franka Panda Cobot Magic
Model SR PSR AS CFS SR PSR AS CFS
Pi0.5 15.8 30.4 83.3 73.7 16.7 42.6 79.5 56.8
Motus 10.8 26.5 69.3 49.4 7.5 31.5 68.5 48.3
LingBot-VLA 3.3 12.1 60.3 54.5 0.0 24.4 47.2 29.3
GROOT N1.6 3.3 11.6 66.3 61.2 1.7 13.7 48.2 38.8
SmolVLA 0.0 6.6 33.3 28.3 0.0 6.3 32.5 27.3
Complete original PDF block S4.F4
Figure 4: Transfer Gap on paired atomic tasks of Pi0.5. Each point pairs a Franka atom with its dual-arm counterpart.

4.4 Failure Analysis

Failed rollouts are labeled motor failure, reference failure, constraint failure, safety abort or timeout. The authors report that motor/reference/timeout account for over 90%, without a complete per-category count table here. The examples distinguish selecting the right object but failing to reorient it from choosing the wrong object despite executing a grasp.

In the paired pouring example, the single arm succeeds while the dual-arm policy fails to hold the receiving bowl, preventing the later pour. This is a concrete dependency failure. Platform comparisons also change embodiments, action sequences and coordination demands, so they do not isolate an arm-count effect.

Complete original PDF block S4.F5
Figure 5: Representative failure cases.
Complete original PDF block S4.F6
Figure 6: Single-arm vs. dual-arm failure comparison on a paired task. The single-arm rollout completes the task, whereas the dual-arm counterpart fails midway.

4.5 Atomic-transfer vs. Single-task Atomic Fine-tuning

Most Franka task pairs favor joint atomic training for π₀.₅, while Cobot Magic is mixed: instruction tasks often benefit but several motor tasks favor specialization. Shared grasp/move/place experience is a plausible explanation, not a separately isolated causal test.

The joint checkpoint sees 1,500 demonstrations across tasks, whereas a single-task checkpoint sees one hundred. The comparison therefore combines transfer with additional data and variation; it is not a matched-total-data verdict that multitask training is universally better. Figure 4 preserves every paired task point.

5 Conclusion

The benchmark shows that five adapted policies still struggle with precise physical operations and held-out skill combinations in this controlled tabletop setting. Its main value is making the training/task relationship and partial-progress measurements explicit, so failures can be inspected beyond a single overall success number.

6 Limitations

The study covers five policies and common tabletop operations, excluding deformables, tools, mobile manipulation and very long-horizon tasks. Manual placement, safety monitoring and score-point annotation remain expensive. Ten physical trials per task provide coarse 10-percentage-point task-level SR increments; uncertainty and training-seed variability are not separately characterized here.

Appendix A Task Design

Tables 3–8 retain all fifty-four task prompts, atom assignments and scene images. Franka uses m/i/x identifiers and Cobot Magic dm/di/dx. Motor tasks include two pick-place instances and two access instances; each track’s seven instruction tasks test attributes, relations, quantities or filtering.

Compositions pair motor operations with multiple constraints—for example, selecting exactly two red cubes, accessing a drawer before relation-based selection, or choosing a receiving bowl before pouring. The dual-arm counterpart may add meaningful requirements beyond coordination: dx9 stacks three specified cubes and dx10 adds color/shape/exclusion. Some listed counterparts, such as dx7/dx8, do not explicitly prescribe two-hand roles. Paired does not mean an otherwise identical task. The original motor-table column heading says Instruction_atom; its entries are motor labels and are retained as printed.

Complete original PDF block A1.T3
Table 3: Single-arm Motor Set (m1–m8). Each task isolates one motor atom under a minimal, unambiguous instruction.
Original selectable table and full-resolution task images
Task_id Prompt Instruction_atom Image
m1 Pick up the cube and place it into the basket. pick_place [Uncaptioned image]
m2 Pick up the ball and place it into the basket. pick_place [Uncaptioned image]
m3 Place the bottle upright on the tray. reorient [Uncaptioned image]
m4 Push the cube into the marked area. push [Uncaptioned image]
m5 Pour the coffee beans from the cup into the bowl. pour [Uncaptioned image]
m6 Open the drawer, then close it. access [Uncaptioned image]
m7 Stack the two cubes. stack [Uncaptioned image]
m8 Open the lid, then close it. access [Uncaptioned image]
Complete original PDF block A1.T4
Table 4: Single-arm Instruction Set (i1–i7). Each task isolates one instruction atom under a pick-and-place carrier motor.
Original selectable table and full-resolution task images
Task_id Prompt Instruction_atom Image
i1 Pick up the blue cube and place it into the basket. color [Uncaptioned image]
i2 Pick up the triangular block and place it into the basket. shape [Uncaptioned image]
i3 Pick up the largest block and place it onto the tray. size [Uncaptioned image]
i4 Pick up the object to the left of the cylinder and place it into the basket. source_relation [Uncaptioned image]
i5 Pick up exactly two blocks and place them into the basket. count [Uncaptioned image]
i6 Pick up the red cube and place it into the right basket. goal_destination [Uncaptioned image]
i7 Place all non-red objects into the basket. exclusion [Uncaptioned image]
Complete original PDF block A1.T5.part1Complete original PDF block A1.T5.part2
Table 5: Single-arm Composition Set (x1–x12). Each held-out task combines one or more motor atoms with two or more instruction atoms.
Original selectable table and full-resolution task images
Task_id Prompt Motor_atom Instruction_atoms Image
x1 Pick up exactly two red cubes and place them into the basket. pick_place color + count [Uncaptioned image]
x2 Pick up the non-red object to the right of the cylinder and place it into the basket. pick_place source_relation + exclusion [Uncaptioned image]
x3 Open the drawer, pick up the red object to the left of the cylinder, place it into the basket, then close the drawer. access + pick_place color + source_relation [Uncaptioned image]
x4 Open the drawer, pick up exactly two blocks, place them into the right basket, then close the drawer. access + pick_place count + goal_destination [Uncaptioned image]
x5 Place the two smallest bottles upright into the tray. reorient count + size [Uncaptioned image]
x6 Place the green bottle upright onto the right tray. reorient color + goal_destination [Uncaptioned image]
x7 Push the non-red block into the right marked area. push exclusion + goal_destination [Uncaptioned image]
x8 Push the triangular block to the left of the cylinder into the marked area. push shape + source_relation [Uncaptioned image]
x9 Stack the smallest red cube on top of the green cube. stack color + size [Uncaptioned image]
x10 Stack the triangular block that is not purple on top of the cube. stack shape + exclusion [Uncaptioned image]
x11 Pour the beans from the bean dish into the small bowl on the right. pour size + goal_destination [Uncaptioned image]
x12 Pour the beans from the brown bean dish on the left into the bowl. pour color + source_relation [Uncaptioned image]
Complete original PDF block A1.T6
Table 6: Dual-arm Motor Set (dm1–dm8). Bimanual counterparts of the single-arm Motor Set, additionally requiring two-arm coordination.
Original selectable table and full-resolution task images
Task_id Prompt Instruction_atom Image
dm1 Pick up the basket with the left hand and place the cube into the basket with the right hand. pick_place [Uncaptioned image]
dm2 Pick up the basket with the left hand and place the ball into the basket with the right hand. pick_place [Uncaptioned image]
dm3 Place the bottles upright on the tray. reorient [Uncaptioned image]
dm4 Push the two cubes into the marked area. push [Uncaptioned image]
dm5 Pick up the bowl with the left hand and pour the coffee beans from the cup into the bowl with the right hand. pour [Uncaptioned image]
dm6 Open the drawer with the left hand, place the block into the drawer with the right hand, then close the drawer with the left hand. access [Uncaptioned image]
dm7 Pick up one cube with the left hand and another cube with the right hand. First place the cube in the right hand at the target position, then stack the cube in the left hand on top. stack [Uncaptioned image]
dm8 Open the box lid with the left hand, place the block into the box with the right hand, then close the box lid with the left hand. access [Uncaptioned image]
Complete original PDF block A1.T7.part1Complete original PDF block A1.T7.part2
Table 7: Dual-arm Instruction Set (di1–di7). Bimanual counterparts of the single-arm Instruction Set, isolating one instruction atom each.
Original selectable table and full-resolution task images
Task_id Prompt Instruction_atom Image
di1 Pick up the basket with the left hand and place the blue cube into the basket with the right hand. color [Uncaptioned image]
di2 Pick up the basket with the left hand and place the triangular block into the basket with the right hand. shape [Uncaptioned image]
di3 Pick up the basket with the left hand and place the largest block into the basket with the right hand. size [Uncaptioned image]
di4 Pick up the basket with the left hand and place the object to the left of the red cylinder into the basket with the right hand. source_relation [Uncaptioned image]
di5 Pick up one cube with each hand and place exactly two cubes into the same basket, leaving the remaining cube outside. count [Uncaptioned image]
di6 Pick up one cube with the left hand and place it into the upper basket. Pick up one cube with the right hand and place it into the lower basket. goal_destination [Uncaptioned image]
di7 Place all geometric objects except the red cube into the basket. exclusion [Uncaptioned image]
Complete original PDF block A1.T8.part1Complete original PDF block A1.T8.part2
Table 8: Dual-arm Composition Set (dx1–dx12). Bimanual held-out compositions combining one or more motor atoms with two or more instruction atoms.
Original selectable table and full-resolution task images
Task_id Prompt Motor_atom Instruction_atoms Image
dx1 Pick up one red cube with each hand and place exactly two red cubes into the basket.
pick_place color + count [Uncaptioned image]
dx2 Pick up the basket with the left hand and place the all non-red objects to the right of the cylinder into the basket with the right hand.
pick_place source_relation + exclusion [Uncaptioned image]
dx3 Open the drawer with the left hand, pick up the red cube to the left of the cylinder with the right hand, place it into the basket, then close the drawer with the left hand.
access + pick_place color + source_relation [Uncaptioned image]
dx4 Open the drawer with the left hand, pick up one block with each hand, place exactly two blocks into the right basket, then close the drawer with the left hand.
access + pick_place count + goal_destination [Uncaptioned image]
dx5 Place the two smallest bottles upright into the tray.
reorient count + size [Uncaptioned image]
dx6 Pick up the two green bottles lying down, one with each hand, and place them upright in the right target area, leaving the bottles of other colors in place.
reorient color + goal_destination [Uncaptioned image]
dx7 Push the block that is not red into the right marked area.
push exclusion + goal_destination [Uncaptioned image]
dx8 Push the triangular block to the left of the cylinder into the marked area.
push shape + source_relation [Uncaptioned image]
dx9 Place the green cube on top of the largest blue cube with your left hand, and put the smallest red cube onto the green cube with your right hand.
stack color + size [Uncaptioned image]
dx10 Pick up the triangular block with the left hand and pick up a cube that is not red with the right hand. First place the cube at the target position, then stack the triangular block on top of it.
stack color + shape + exclusion [Uncaptioned image]
dx11 Pick up the smaller of the two bowls with the left hand, pick up the bean dish containing beans with the right hand, and pour the beans from the bean dish into that smaller bowl.
pour size + goal_destination [Uncaptioned image]
dx12 Pick up the bowl with the left hand, pick up the brown bean dish with the right hand, and pour the beans from the brown bean dish into the bowl.
pour color + source_relation [Uncaptioned image]

Appendix B Physical Test Seed Placement

Before every rollout, the operator aligns objects with a semi-transparent reference image in the camera view. Figure 7 preserves both front and side examples. The method improves consistency across policies without claiming exact repeated poses or eliminating physical reset variability.

Complete original PDF block A2.F7
Figure 7: Mask-guided object placement. The reference scene of a physical test seed (overlaid as a mask) lets the operator reproduce the same initial object positions across policies and evaluation runs.

Appendix C Robustness to Language and Visual Perturbations

Robustness reuses the atomic-transfer checkpoint without adaptation and selects tasks it already performs well on. The reported experiment is specifically π₀.₅ on Cobot Magic, four atomic and two compositional tasks; it is not an all-model or all-task robustness survey.

Visual perturbations (Perturbed-Vis).

The visual change replaces the white tabletop cloth with brown, retaining objects, arrangement, lighting, task rubric and physical seeds. This is one controlled background shift, not a test of broad visual distribution changes.

Language perturbations (Perturbed-Lan).

One authored paraphrase per selected task changes verbs, role phrasing or sentence structure while preserving atom-defining words, left/right roles and temporal order. Table 9 keeps every original/paraphrased pair. Because key constraint tokens remain fixed, the test does not cover arbitrary language reformulation or new semantics.

Complete original PDF block A3.T9
Table 9: Original and language-perturbed (Perturbed-Lan) instructions for the evaluated tasks.
Original selectable table
Task Original prompt Perturbed-Lan prompt
di1 Pick up the basket with the left hand and place the blue cube into the basket with the right hand. Use your left hand to hold the basket, and use your right hand to drop the blue cube into it.
di2 Pick up the basket with the left hand and place the triangular block into the basket with the right hand. Hold the basket with the left hand; with the right hand, put the triangular block inside.
dm2 Pick up the basket with the left hand and place the ball into the basket with the right hand. Grasp the basket using the left hand, then drop the ball into it using the right hand.
dm6 Open the drawer with the left hand, place the block into the drawer with the right hand, then close the drawer with the left hand. Pull the drawer open with the left hand, put the block inside with the right hand, then push the drawer shut with the left hand.
dx1 Pick up one red cube with each hand and place exactly two red cubes into the basket. Use each hand to grasp one red cube, then place exactly two red cubes into the basket.
dx12 Pick up the bowl with the left hand, pick up the brown bean dish with the right hand, and pour the beans from the brown bean dish into the bowl. Lift the bowl with the left hand and the brown bean dish with the right hand, then tip the beans from the brown bean dish into the bowl.

Results.

The four atomic tasks remain at 90–100%. Composition dx1 falls from 60% to 40% under both shifts; dx12 falls from 50% to 40% with the cloth change and 20% with the paraphrase. Across the six tasks, mean SR is 78.3% baseline, 76.7% visual and 71.7% language.

These are small samples with ten trials per condition, selected partly for prior success. They suggest brittleness in the two tested compositions, but do not prove that the atomic representation is generally disentangled or that paraphrasing is always a stronger stress test. Improvements on some atomic trials partly offset composition losses in the six-task average.

Complete original PDF block A3.F8
Figure 8: Pi0.5 success rate on Cobot Magic under the baseline atomic-transfer protocol, Perturbed-Vis (brown tablecloth), and Perturbed-Lan (paraphrased instruction). The right-most group is the mean SR across the six tasks.

Appendix D SFT Configurations for Models in ATOM-Bench

All runs use H100 GPUs and official model code/recipes where possible. π₀.₅, LingBot-VLA, GR00T and Motus use full fine-tuning; SmolVLA uses LoRA. AdamW learning rates, schedules, batch sizes and epochs vary across models and, for several models, across platforms; Table 10 retains every setting.

Examples include three epochs for π₀.₅ on both platforms, three/six for GR00T and nine/five for SmolVLA. This is a practical recipe-based comparison, not equalized optimization compute. Joint adaptation uses 1,500 demonstrations per platform; the π₀.₅ specialization baseline uses one hundred per checkpoint.

Complete original PDF block A4.T10
Table 10: Per-model SFT configurations in atomic-transfer.
Original selectable table
Model Base ckpt. Mode Optimizer LR Batch Steps
Franka Panda
Pi0.5 pi05_base full AdamW 2.0e-5 (cosine, warmup steps 1000) 256 3 epochs
LingBot-VLA lingbot-vla-4b full AdamW 2.5e-5 (cosine, warmup ratio 0.25) 128 3 epochs
GROOT N1.6 GR00T N1.6 full AdamW 1.0e-4 (cosine, warmup ratio 0.05) 256 3 epochs
SmolVLA smolvla_base LoRA AdamW 1.0e-4 (cosine, warmup steps 1000) 256 9 epochs
Motus Motus full AdamW 2.0e-5 (linear, warmup steps 1000) 64 6 epochs
Cobot Magic
Pi0.5 pi05_base full AdamW 2.0e-5 (cosine, warmup steps 1000) 256 3 epochs
LingBot-VLA lingbot-vla-4b full AdamW 5.0e-5 (constant) 256 3 epochs
GROOT N1.6 GR00T N1.6 full AdamW 1.0e-4 (cosine, warmup ratio 0.05) 256 6 epochs
SmolVLA smolvla_base LoRA AdamW 1.0e-4 (cosine, warmup steps 1000) 256 5 epochs
Motus Motus full AdamW 2.0e-5 (linear, warmup steps 1000) 64 5 epochs

Appendix E Atom Taxonomy and Design Rationale

E.1 Motor Atoms

The six motor families draw on Meta-World, FMB and manipulation motion taxonomies. Stacking is separated from transport because relative pose and stable contact matter; access groups opening, maintaining and closing articulated containers. These functional categories are useful abstractions rather than irreducible elementary motions.

Complete original PDF block A5.T11
Table 11: Motor atoms in ATOM-Bench. Each atom summarizes a recurring physical operation family used for real-robot adaptation and held-out compositional evaluation.
Original selectable table
Motor Atom Definition Design Rationale
pick_place Grasp an object and place it into a target container or region. Foundational tabletop operation and default carrier motor for many instruction atoms.
reorient Change an object’s pose or orientation before completing placement, such as placing a bottle upright. Tests object pose control beyond basic transport.
push Move an object along a support surface to a target region without necessarily grasping it. Captures canonical non-prehensile tabletop manipulation.
stack Place one object on top of another object. Tests relative-pose precision and contact stability.
pour Transfer granular or liquid-like contents from one container to another. Tests functional container use and material transfer.
access Open, maintain, or close an articulated container, such as a drawer, box, or lid. Tests articulated-object interaction and phase transitions.

E.2 Instruction Atoms

Instruction factors draw on ReferItGame, unambiguous object descriptions, CLEVR and embodied benchmarks. Color, shape, size, spatial source relation and exclusion select what to manipulate; count specifies how many; goal_destination binds where to deliver it. Table 12 preserves the exact definitions and design rationale for all seven.

Complete original PDF block A5.T12
Table 12: Instruction atoms in ATOM-Bench. These atoms summarize recurring semantic and logical constraints in language-conditioned manipulation.
Original selectable table
Instruction Atom Definition Design Rationale
color Select the target object according to color. Basic attribute for object-level referring expressions.
shape Select the target object according to geometric shape. Stable visual attribute for tabletop object selection.
size Select the target object according to relative size, such as the largest or smallest object. Tests comparative attribute grounding.
count Manipulate an exact number of objects. Tests set-level reasoning and multi-object selection.
exclusion Select objects by excluding a specified attribute or category. Tests logical filtering over candidate objects.
source_relation Select the target object according to its spatial relation to another reference object. Tests relation-based object reference.
goal_destination Bind the action endpoint to one among multiple candidate destinations or target regions. Tests binding between instruction and action endpoint.
Concept demonstration

Open image ↗