World models help agents assess the consequences of their actions before acting. Many real systems, however, involve multiple actors who make their own decisions, can evolve even when the agent takes no action, and are only partially observed through event logs. We formalize these systems as event worlds and ask whether large language models can predict future observations from the log and the agent's proposed actions. We build complementary benchmarks from human Diplomacy games and real Polymarket trades. Our hierarchical framework compares prediction with and without other actors' actions, then tests multi-step rollouts, predictions for actions never taken, and usefulness to agents. Fine-tuning and more data improve prediction when all actors' recorded actions are given. With only the agent's action, models struggle when other actors strongly affect the outcome, and larger models and longer histories help little. This gap also limits multi-step prediction and agent decisions, with models overestimating plan success. Yet imagined experience and initialization from world models can still improve policy learning. Anticipating other actors' actions remains a central challenge for event world modeling.
Figures & tables
Figure 1: From single-actor environments to multi-actor event worlds. Ht , A , and O denote the event log, actions, and observations.
Diplomacy
Polymarket
World type
board game
market
Actors I
2 or 7 players
anonymous traders
Observation O
the board
trading statistics
Hidden part X
players’ intentions
order book, beliefs
Actions A
move, hold, support
buy or sell
Horizon Δ
one phase
one hour
Table 1: The two event world benchmarks.
Diplomacy
Polymarket
Setting
Model
EM
EM high
EM low
F1
Acc
Acc high
Acc low
CRPS ↓
Task J (joint action)
Zero-shot
GPT-6-Luna
29.1
3.4
55.2
90.8
43.0
36.8
48.0
0.42
GPT-6-Sol
40.5
10.1
71.4
96.3
54.6
48.7
59.5
0.55
SFT
Qwen3-1.7B
59.7
23.6
96.4
96.9
66.1
62.5
69.0
0.29
Qwen3-8B
65.1
30.8
100.0
97.7
65.5
60.5
69.5
0.29
Table 2: Single-step prediction (%, CRPS in cents) on decision points shared by all models, with every actor’s action given (Task J, Q1) or only the agent’s (Task S, Q2). EM high /Acc high and EM low /Acc low are restricted to high- and low-interaction transitions. Fine-tuned models (SFT) are evaluated zero-shot. Few-shot results are in Appendix E.1 . Bold marks the best prompted or fine-tuned LLM per column, baselines excluded.
Figure 2: The evaluation framework. The agent is orange, actors whose actions are hidden are dashed, and the dashed branch is an action never taken.
Figure 3: Scaling in training data and model size.
Figure 4: (a) Predicted probability that the Polymarket price does not change against the observed frequency. (b) Gain from knowing the other actors’ actions (Task J minus Task S) when the agent acts and when it is idle.
Figure 5: Task S against history length K . The full sweep is in Appendix F.1 .
Diplomacy
Polymarket
Task
Model
F1 closed
F1 open
Acc closed
Acc direct
J
Qwen3-1.7B
91.9
97.5
60.2
64.4
Qwen3-8B
94.1
97.9
61.2
63.8
GPT-6-Luna
65.7
91.7
57.4
35.6
GPT-6-Sol
83.4
96.4
57.4
55.6
S
Qwen3-1.7B
55.1
86.0
55.6
58.6
Table 3: Long-horizon rollout results. Diplomacy: F1 at step 5 when the model is fed its own predictions (closed loop) or the true board (open loop). Polymarket: Acc over 24 hours from recursive price prediction with logged trading activity (closed) or a direct forecast. Full table in Appendix G.1 .
F1
Spurious ↓
Model
S
J
S
J
GPT-6-Luna
47.7
30.1
45.9
89.7
GPT-6-Sol
56.5
57.9
40.4
83.3
Qwen3-1.7B
44.0
73.0
51.5
33.1
Qwen3-8B
44.3
77.3
59.6
29.4
Table 4: Results after replacing the agent’s orders in Diplomacy (%). F1 compares predicted and true changes to units and supply centers. Spurious is the rate of predicting a change when original and modified orders produce the same board. More detailed results are reported in Appendix G.2 .
Figure 6: Success of plans made inside an LLM, judged inside the LLM and when replayed in the engine, against plans made in the engine.
Figure 7: Evaluation of candidate actions by LLMs in Diplomacy. (a) Gain in supply centers over B=1 . (b) Predicted and realized gain of the chosen candidate.
Training data (Qwen3-1.7B policy)
Δ SC
Success (%)
Behavior policy (DipNet)
0.24
61.0
Real 1k
− 0.58
47.5
Imagined 1k
− 0.62
43.0
Real 1k + imagined 1k
− 0.29
53.5
Real 2k
− 0.23
54.0
Imagined 2k
− 0.48
47.5
Table 5: Policies behavior-cloned on real or imagined games, or initialized from a world model. Δ SC: mean change in supply centers over 10 phases.
Line of work
Main prediction or evaluation target
Relation to our study
Text, web, tool world models
Observations after an agent’s actions
We extend evaluation to logs with multiple actors whose future actions are unknown.
Visual and driving world models
Frames, scene states, or joint trajectories
Event logs can record actions and observations at irregular times.
Multi-agent benchmarks
Agents’ collaboration and competition
We evaluate world predictions from human game records and real market logs, then their usefulness to an agent.
Models of other actors
Other participants’ actions or beliefs
These predictions are one part of predicting the world’s future observations.
Event forecasting
Future events from records or news
Our predictions additionally condition on the agent’s proposed actions.
Offline model-based RL
Dynamics and policies learned from logged transitions
We examine the difficulty introduced by unknown actions of other actors.
Table 6: Related work compared by its prediction or evaluation target and its relation to event world modeling. Representative studies are discussed in Appendix A .
Appendix figures & tables20 assets
Supplementary material from the paper’s appendix.
Appendix
Diplomacy
Polymarket
clock
phase counter
physical time in seconds
horizon Δ
one phase
chosen; 1 h (also 5 min, 24 h)
actors I
7 (or 2) players
taker wallets
action At+1i
one order per unit; hold if none
0 to many orders (time, direction, size)
log Ht
boards, orders, results
orders in earlier blocks
observation Ot
units, centers
prices, flow, depth proxies
Appendix
Table 7: How the general definition maps to the two worlds. Changed actions have engine-verified outcomes in Diplomacy. Polymarket provides an observational response check (Appendix G.3 ).
Question
Experiment
Evaluation reference
Diplomacy
Polymarket
Sec.
Q1: every action given
Does the model predict the next observation when every actor’s action is given?
the log
next board from all players’ orders
price move over the window from all orders
6.1
Does more data or a larger model help, and is pretraining needed?
the log
5k to 320k transitions; 0.6B to 8B parameters; training from scratch
same
6.1
Does the model know how much the world changes as time passes?
the log
horizons of 5 min, 1 h, and 24 h
6.1
Does what the model learns hold when the world shifts?
the log
two-player games, unseen in training
markets that open after the training period
6.1
Q2: only the agent’s action
How much does knowing the other actors’ actions improve prediction?
the log
Task J vs. Task S, by interaction and when the agent issues no orders
Task J vs. Task S, including windows where the agent does nothing
6.2
Do history, predicting the other actors’ behavior, or scale close the gap?
the log
up to 10 previous phases; factored variant, its orders resolved by the engine; data and model size
up to 10 previous hours; factored variant
6.2
Appendix
Table 8: The evaluation framework of Section 6 , one experiment per row.
Seed
J
S
SF
SF+engine
Top-1
42
59.7
20.1
22.2
23.1
47.6
43
59.6
19.6
21.5
22.1
46.8
44
60.7
19.9
23.1
23.4
48.1
Mean (sd)
60.0 (0.6)
19.8 (0.3)
22.2 (0.8)
22.9 (0.7)
47.5 (0.7)
Appendix
Table 9: Qwen3-1.7B fine-tuned on 20k Diplomacy transitions with three training seeds (%, EM except Top-1). SF first predicts the other players’ orders, then the next board, and SF+engine resolves its predicted orders in the engine (Appendix F.2 ). Top-1 is the share of the other players’ orders that the SF model predicts correctly, computed per sample and averaged over samples, with samples in which the other players issue no orders counted as correct. Per-order accuracy (Table 18 ) instead pools all units.
K=0
K=3
Setting
Model
EM
EM high
EM low
F1
EM
EM high
EM low
F1
Task J (joint action)
Zero-shot
GPT-6-Luna
29.1
3.4
55.2
90.8
27.8
3.6
52.4
89.8
GPT-6-Sol
40.5
10.1
71.4
96.3
53.3
18.8
88.3
96.5
Few-shot (3-shot)
GPT-6-Luna
27.6
1.6
54.0
89.8
23.2
1.4
45.4
89.1
GPT-6-Sol
43.8
12.1
76.0
96.0
49.7
14.7
85.3
96.6
Appendix
Table 10: Single-step prediction in seven-player Diplomacy (%) on the shared subset (1,000 transitions, 2,000 Task S samples). K is the number of previous transitions given as history. EM high and EM low are restricted to high- and low-interaction transitions. Fine-tuned models are evaluated zero-shot, and baselines use no history. In the “+ engine” rows, the named policy supplies the other players’ orders and the engine resolves the joint action (Appendix D.3 ). Bold marks the best prompted or fine-tuned LLM per column, baselines excluded.
K=0
K=3
Setting
Model
Acc
Acc high
Acc low
Dir
CRPS
Acc
Acc high
Acc low
Dir
CRPS
Task J (joint action)
Zero-shot
GPT-6-Luna
43.0
36.8
48.0
56.2
0.42
45.9
40.6
50.2
57.2
0.45
GPT-6-Sol
54.6
48.7
59.5
56.5
0.55
55.4
51.3
58.8
57.6
0.63
Few-shot (3-shot)
GPT-6-Luna
44.1
41.1
46.5
56.4
0.41
41.6
37.1
45.3
56.6
0.39
GPT-6-Sol
52.7
52.3
53.1
58.4
0.49
52.5
51.3
53.5
59.0
0.43
Appendix
Table 11: Single-step prediction in Polymarket ( Δ=1 h, accuracies in %) on the shared subset (1,333 decision points). K is the number of previous hours given as history. Acc high and Acc low are restricted to high- and low-interaction windows. Dir is the direction accuracy on windows in which the mid-price changes. CRPS is in cents (lower is better). Fine-tuned models are evaluated zero-shot, and baselines use no history. Bold marks the best prompted or fine-tuned LLM per column, baselines excluded.
Diplomacy (EM)
Polymarket (Acc)
Axis
Setting
J
S
J
S
Data (1.7B)
5k
48.9 (11.3)
16.2 (0.8)
62.0 (56.5)
56.6 (53.2)
20k
59.7 (23.6)
20.1 (1.6)
66.1 (62.5)
59.9 (55.2)
80k
74.7 (49.8)
23.0 (1.9)
69.5 (66.7)
61.8 (55.4)
320k
88.4 (77.0)
25.4 (1.9)
71.6 (69.1)
64.9 (58.9)
Size (20k)
0.6B
57.6 (19.2)
20.5 (1.6)
63.3 (59.2)
58.1 (53.8)
Appendix
Table 12: Scaling the fine-tuned Qwen3 models in training data and model size (%, shared subsets). Values on high-interaction transitions are in parentheses.
Diplomacy (EM)
Polymarket (Acc)
Data
Initialization
J
J high
J low
J, 2p
S
J
J, later
S
20k
pretrained
59.7
23.6
96.4
61.1
20.1
66.1
67.4
59.9
random
32.0
4.6
59.9
0.3
10.2
60.8
59.6
57.8
320k
pretrained
88.4
77.0
100.0
88.4
25.4
71.6
74.0
64.9
random
66.2
32.9
100.0
11.0
24.1
67.5
65.4
61.7
Appendix
Table 13: The Qwen3-1.7B architecture, fine-tuned from the pretrained checkpoint or trained from random initialization (%, K=0 ). J high , J low : high- and low-interaction transitions. J, 2p: Task J on the two-player common evaluation set. J, later: Task J on 1,333 decision points of the later markets. The other columns use the shared subsets.
Qwen3-1.7B
Qwen3-8B
Baseline
Split
All
High
All
High
All
High
Diplomacy, Task J
7p
60.9
22.9
65.7
30.7
52.0
4.0
2p
61.1
38.6
69.6
53.0
53.1
23.9
Diplomacy, Task S
7p
19.7
1.2
19.8
1.2
7.3
1.5
Appendix
Table 14: Robustness on the common evaluation sets of Appendix B.3 (EM in Diplomacy, Acc in Polymarket, %). High: high-interaction transitions. 7p: seven-player games. 2p: two-player games, unseen in training. Test: the in-distribution test split. Later: the later-market split. Baselines: orders applied literally in Diplomacy (in Task S only the agent’s own), LightGBM in Polymarket.
Outcome known
Acc, test → later
Model
Test
Later
Task J
Task S
GPT-6-Luna
81.9
1.6
43.0 → 46.7
46.1 → 48.5
GPT-6-Sol
68.1
0.0
54.6 → 57.5
48.5 → 52.6
Qwen3-1.7B (SFT)
–
–
66.1 → 67.4
59.9 → 61.4
Qwen3-8B (SFT)
–
–
65.5 → 70.7
60.2 → 63.1
LightGBM
–
–
68.8 → 67.1
61.4 → 61.8
Appendix
Table 15: Left: share (%) of resolved markets whose outcome each GPT-6 model says it knows (238 test and 440 later markets). Among test markets whose outcomes the models say they know, accuracy is 97.9% for GPT-6-Luna and 98.8% for GPT-6-Sol. Right: Task J and Task S Acc (%) on the 1,333-point subsets of the in-distribution test split and the later-market split.
Diplomacy (EM)
Polymarket (Acc)
Model
Task
K=0
K=1
K=3
K=6
K=10
K=0
K=1
K=3
K=6
K=10
Qwen3-1.7B (SFT)
J
58.9
59.4
59.6
60.1
59.7
65.5
66.5
66.8
66.2
66.5
S
19.9
22.2
21.9
21.1
20.8
60.5
60.8
61.9
61.8
61.1
S, low
38.8
43.6
42.8
41.2
40.7
63.8
63.8
65.0
64.9
63.7
S, high
1.3
1.2
1.3
1.2
1.2
56.5
57.2
58.0
58.0
58.0
GPT-6-Sol
J
40.5
47.4
53.3
54.5
51.7
54.6
56.3
55.4
54.1
54.8
Appendix
Table 16: History-length sweep on the shared subsets (EM in Diplomacy and Acc in Polymarket, in %). All models are evaluated zero-shot. K is the number of previous transitions or hours in the input. “S, low” and “S, high” restrict Task S to low- and high-interaction transitions.
Diplomacy (EM)
Polymarket (Acc)
Model
J
S
SF
SF+engine
Share
J
S
SF
Share
GPT-6-Luna (zero-shot)
29.1
6.7
9.3
17.2
0.24
43.0
46.1
47.6
− 0.06
GPT-6-Sol (zero-shot)
40.5
7.5
11.9
17.1
0.36
54.6
48.5
52.6
0.12
Qwen3-1.7B (SFT)
59.7
20.1
22.2
23.1
0.50
66.1
59.9
60.4
0.16
Qwen3-8B (SFT)
65.1
20.8
22.9
23.4
0.56
65.5
60.2
–
0.13
DipNet policy + engine
–
–
–
24.7
–
–
–
–
–
Appendix
Table 17: Prediction with and without the other actors’ actions. SF adds predictions of their behavior to the Task S output, and SF+engine resolves its predicted orders in the engine. Share =(ErrS−ErrJ)/ErrS , with Err the error rate, is the relative reduction in error rate from Task S to Task J.
Model
All units
Interacting
EM high
Most frequent order
22.6
16.2
0.1
GPT-6-Luna (zero-shot)
16.8
10.9
1.5
GPT-6-Sol (zero-shot)
20.7
16.5
0.1
Qwen3-1.7B (SFT, 3 seeds)
29.6
25.5
1.9
Qwen3-8B (SFT)
30.7
26.6
1.8
Qwen3-1.7B (SFT), K=3
36.1
31.9
1.8
Appendix
Table 18: Per-order accuracy (%) of the SF models and two baselines on the other players’ orders in the 2,000 Task S samples of the shared subset, over all units and over interacting units. EM high is the high-interaction EM after the engine resolves the predicted orders.
Task S EM
Others’ orders
Model
none
messages
blank
SF
SF + msg.
Qwen3-1.7B, press-tuned
18.3
18.2
17.6
47.0
47.9
GPT-6-Sol
8.9
11.1
10.2
31.8
32.9
GPT-6-Luna
6.8
6.9
7.1
26.3
26.2
Appendix
Table 19: Task S in press games (%, prompted models zero-shot). EM is given without the agent’s messages, with them, and with their text blanked. Others’ orders gives SF Top-1 (Table 9 ), without and with messages.
Retreat
Adjustment
Both
Model
idle
acting
idle
acting
idle
acting
GPT-6-Luna (zero-shot)
23.0
9.0
77.8
70.0
38.6
49.1
GPT-6-Sol (zero-shot)
35.8
17.9
92.4
91.8
51.9
66.5
Qwen3-1.7B (SFT)
45.6
18.7
80.7
82.5
55.6
60.6
Qwen3-8B (SFT)
47.7
23.9
85.4
82.5
58.4
62.4
Appendix
Table 20: Diplomacy, Task J minus Task S EM (points) when the agent issues no orders (idle) or some (acting), by phase type. Retreat phases have 430 idle and 134 acting samples, and adjustment phases 171 and 257.
Diplomacy, 5-step chains
Polymarket, 24 hours
Task
Model
EM closed
EM open
F1 closed
F1 open
Acc closed
Acc direct
J
Qwen3-1.7B (SFT)
5.6
65.4
91.9
97.5
60.2
64.4
J
Qwen3-8B (SFT)
13.2
68.8
94.1
97.9
61.2
63.8
J
GPT-6-Luna
0.0
42.2
65.7
91.7
57.4
35.6
J
GPT-6-Sol
0.2
57.2
83.4
96.4
57.4
55.6
S
Qwen3-1.7B (SFT)
0.0
22.0
55.1
86.0
55.6
58.6
Appendix
Table 21: Multi-step prediction (%), extending Table 3 . Diplomacy scores the board at step 5 with predicted boards (closed loop) or recorded boards (open loop) as inputs. Polymarket recursively predicts prices with logged trading activity (closed) or forecasts 24 hours directly. Always predicting a fall gives 49.4% Acc.
Task S form (other actors imagined)
Task J form (other actors given)
Model
F1
Insensitive ↓
Spurious ↓
Size pred/ref
F1
Insensitive ↓
Spurious ↓
Size pred/ref
GPT-6-Luna (zero-shot)
47.7
6.0
45.9
3.4/3.9
30.1
0.1
89.7
9.5/3.9
GPT-6-Sol (zero-shot)
56.5
4.8
40.4
3.9/3.9
57.9
0.3
83.3
7.6/3.9
Qwen3-1.7B (SFT)
44.0
6.0
51.5
5.4/3.9
73.0
3.2
33.1
4.3/3.9
Qwen3-8B (SFT)
44.3
5.6
59.6
5.6/3.9
77.3
2.9
29.4
4.3/3.9
Appendix
Table 22: Diplomacy action changes, extending Table 4 , on 5,040 order-set evaluations from the shared subset (%, except Size). F1 and Insensitive cover alternatives that change the next board. Spurious covers alternatives that do not, excluding controls that repeat the original orders. Size reports the mean predicted and true numbers of changes to units and centers across all evaluations.
Model
Sign
Insensitive ↓
Magnitude
Monotonicity
GPT-6-Luna (zero-shot)
93.6
6.9
1.74
0.20
GPT-6-Sol (zero-shot)
84.1
15.4
2.94
0.21
Qwen3-1.7B (SFT)
52.2
41.4
0.77
0.12
Qwen3-8B (SFT)
52.3
34.2
0.93
− 0.02
Appendix
Table 23: Polymarket price-response checks in Task S (%, except Magnitude and Monotonicity). The first three columns reverse all of the agent’s orders. Sign measures agreement with the nonzero reference response, counting a zero prediction as a miss. Insensitive counts predicted responses smaller than 0.0005 in absolute value. Magnitude is the ratio of mean absolute predicted to reference responses. Monotonicity instead removes the orders and computes Spearman correlation between relative order volume and the predicted effect of including them, with its sign oriented by the anchor order’s buy or sell direction.
Task
Model
WM
W2R
CR
J
Qwen3-1.7B (SFT)
55.8
50.8
0.78
J
Qwen3-8B (SFT)
55.6
50.6
0.78
J
GPT-6-Luna
18.2
34.2
0.53
J
GPT-6-Sol
41.0
41.4
0.64
S
Qwen3-1.7B (SFT)
82.8
41.2
0.63
S
Qwen3-8B (SFT)
82.8
42.2
0.65
Appendix
Table 24: Plan consistency on 500 starts. WM is success inside the world model, and W2R is success when its plan is replayed in the engine (both in %). CR is W2R/Real, where Real =65.0% is success when the agent acts directly in the engine.
Agent
Task
Real
WM
W2R
CR
Supervised
J
65.0
55.8
50.8
0.78
S
65.0
82.8
41.2
0.63
RL
J
67.4
55.6
50.2
0.74
S
67.4
85.2
46.4
0.69
Appendix
Table 25: Plan consistency of the Qwen3-1.7B world model (20k) with a supervised or an RL-trained DipNet agent (success rates in %, 500 starts).
Training data
Real
Imagined
Pairs
Real 2k
2,000
0
12,974
Real 1k + imagined 1k
1,000
1,000
12,580
Imagined 2k
0
2,000
12,271
Appendix
Table 26: Successful real and imagined games and their board and order pairs for behavior cloning.
World models offer a principled way to equip long-horizon LLM agents with foresight: predictions of action consequences before execution. However, unreliable foresight can be ignored, misused, or even degrade downstream decision-making. In this paper, we introduce WorldEvolver, a self-evolving world model framework that revises its deployment-time context while keeping the downstream agent and all model parameters frozen. WorldEvolver integrates three modules: (i) Episodic Memory, which exploits real action transitions through retrieval-based simulation; (ii) Semantic Memory, which extracts persistent heuristic rules from prediction-observation mismatches; and (iii) Selective Foresight, which filters low-confidence predictions before integrating them into agent reasoning context. We evaluate WorldEvolver on ALFWorld and ScienceWorld, measuring world model prediction accuracy on Word2World and downstream agent success rate on AgentBoard. Extensive experiments show that WorldEvolver achieves the highest prediction accuracy across three backbones and leads other world model baselines on downstream agent success rate, demonstrating that test-time memory revision enhances both predictive fidelity and planning performance.
Xuan Zhang, Wenxuan Zhang, See-Kiong Ng +1
National University of Singapore · Singapore University of Technology and Design · Singapore Management University
How do LLM agents come to both understand environments they act in and master tasks set within them? Through controlled experiments combining world-model training (next-state prediction) and policy training (reward maximization), we investigate this question. We dissect the resulting models through their additive parameter updates. Geometrically, we find effective world-model updates are low-rank and share an input-feature subspace with policy updates while writing to nearly orthogonal output directions, whether trained separately or sequentially. However, we find that, in projection interventions, the sequential update induces more robustness than separate policy RL when removing the world model's leading input directions, suggesting that it has learned alternative input pathways. Behaviorally, we find the sequentially trained agent explores a wider range of states and actions. Based on this, we ask: does policy training preserve world knowledge as well as it could? We probe this with training-free merging built on the geometrically motivated input basis plus an online world-model loss during policy RL, and show both improve over the untreated baseline. Our findings suggest world knowledge and task-directed ability can be learned in geometrically complementary forms, and that future post-training pipelines should consider how best to engineer the interface between them.
Large language models (LLMs) are being used as policies for autonomous decision-making and planning in many domains. Despite their strong reasoning capabilities, LLMs struggle with long-horizon tasks, especially under partial observability. World models are a promising way to enhance policy performance, both during training and inference. During inference, agents currently use world models to simulate the consequences of candidate actions before choosing an action, which can improve decision-making. However, we argue that simulation alone is an incomplete interface for decision-making under partial observability: simulation does not adequately capture uncertainty about the current state, which agents may need for accurate decision-making. We address this limitation with Belief-Based World Models (BB-WMs), which maintain a belief that LLMs can query to access information on what is known and uncertain about the current state. Before developing methods to learn accurate BB-WMs, this paper focuses on a more fundamental question: does exposing a world model's belief directly to an LLM policy improve decision-making? Our results show that giving LLM agents access to beliefs improves task performance under partial observability, while remaining complementary to existing simulation-based world models. Code: https://github.com/skumar-ml/belief-world-models.