Authors: Ayse Sila Okcu, Weisheng Wang, Tongmeng Xie, Niall Dalton, Angelos Nalmpantis, Raghu Arghal, Mario Giulianelli
Organizations: University of Cambridge · Utrecht University · University of York · University College London, University of London · TKH AI Technology · University of Pennsylvania
Language model (LM) agents can fail for different reasons, requiring different explanations and interventions. We study how state beliefs, action values, and action preferences interact as LM agents reason and commit to decisions in text-rendered grid worlds. We estimate task-relevant beliefs using report-based and representation-based probes, and action values using inverse reinforcement learning. We find that task-relevant state and value information is often represented more reliably than it is reflected in behaviour. Tracking action preferences throughout reasoning further shows that decisions sharpen around a commitment point and reported belief uncertainty decreases near commitment, although individual belief changes do not consistently explain shifts in action preference. Finally, forced actions follow reasoning prefixes from other decisions when transplanted, whereas the tested reasoning-window activation edits provide limited control over the action. Together, these results suggest that some failures arise not from missing task-relevant information, but from how available information is translated into action. Our measurements distinguish information availability from belief-action consistency without by themselves identifying the cause of each failure. Our code is available at: https://anonymous.4open.science/r/belief-action-gap-F745/
Figures & tables
Figure 1: Example DoorKey environment shown in text and spatial form. The language model receives the text-rendered grid in (a) , while (b) visualises the same underlying environment used to define states, transitions, and planner-optimal actions. A denotes the agent, K the key, D the door, G the goal, and filled cells denote walls.
All Steps
All Probes Correct
Probe Held-out
Quantity
Pre
Post
Pre
Post
Pre
Post
Expected local regret Δ
0.681
0.582
0.676
0.554
0.700
0.670
Belief–action gap R
0.467
0.268
0.514
0.288
0.316
0.199
Value of perfect state information I
0.214
0.314
0.163
0.266
0.384
0.472
Decoded Belief entropy H
1.494
2.085
1.106
1.811
2.788
3.020
Table 1: Mean expected local regret Δ , belief–action gap R , expected value of perfect state information I , and decoded belief entropy H before and after reasoning. The expected local regret decomposes exactly as Δ=R+I .
Figure 2: Example expected local regret under the pre-reasoning and post-reasoning conditions. Cell numbers show the probe-estimated location probabilities, with shading proportional to probability mass. Arrows mark the successor state preferred by the agent-supervised value model and the action actually emitted by the agent. In this example, reasoning increases the decoded belief entropy ( H:0.261→1.572 ) while reducing the expected local regret ( Δ:2.386→0.205 ).
Figure 3: Evolution of action preference over reasoning (95 trajectories, 1,276 decisions, 98,310 evaluated prefixes). left. The mean probability assigned to the eventual action and inverse normalised action entropy as reasoning progresses, together with the same quantities before the first decisional action mention. right. The probability assigned to the eventual action separately by action. Bands are 95% trajectory-clustered bootstrap intervals.
Figure 4: Decodability and intervention effectiveness at the reasoning-window site. Left: decodability of action decisiveness against positional and length-based baselines. Right: effectiveness of the tested activation interventions, distinguishing any change in the forced action from changes to the intended target action.
Appendix figures & tables13 assets
Supplementary material from the paper’s appendix.
Appendix
Query family
State–question pairs
Greedy
Ten-sample modal
Wall
132
93.9%
97.0%
Goal adjacency
132
100.0%
100.0%
Door adjacency
132
99.2%
100.0%
Key adjacency
132
67.4%
97.0%
Appendix
Table 2: Report accuracy on the 33-state failure-mode set. Each family contains four directional questions per state. The modal result uses ten samples per question at temperature 0.7.
Direction
Report-based
Probe (Pre-reasoning)
Probe (Post-reasoning)
left
100.0%
73.3%
73.3%
right
100.0%
86.7%
93.3%
up
100.0%
100.0%
93.3%
down
100.0%
80.0%
93.3%
Appendix
Table 3: Accuracy of report-based and representation-based estimates of whether the adjacent cell in each direction is blocked, evaluated on the same 15 decision states.
Metric
Pre-reasoning
Post-reasoning
has_key accuracy
100.0%
89.5%
door_open accuracy
100.0%
88.6%
Location top-1 accuracy
34.8%
13.1%
Top-1 within Manhattan distance 1
95.5%
60.8%
Top-1 within Manhattan distance 2
99.0%
83.5%
Mean probability mass within distance 1
93.1%
59.2%
Appendix
Table 4: Expanded representation-based state decoding on held-out environment states. Location top-1 accuracy requires an exact match. “Top-1 within Manhattan distance k ” measures whether the most probable decoded location lies within distance k of the true location. “Mean probability mass within distance k ” is the probability assigned to all locations within that distance of the true position.
Figure 5: Example learned state values for planner-supervised IRL (left) and agent-supervised IRL (right), using post-reasoning activation-derived state representations. Darker cells indicate larger learned values. Because the one-step objective identifies relative successor-state preferences, absolute values should not be interpreted as goal distances; the comparison is intended to visualise differences in the learned state-preference structure.
Figure 6: Additional commitment timing and stability results. (a) Cumulative distributions of the first decisional action mention, the action-commitment prefix, and the point after which the eventual-action probability remains above 0.8. (b) Left: probability that the preferred action changes at the next evaluated prefix as reasoning progresses. Right: probability of at least one later preference switch as a function of the current probability assigned to the eventual action.
Figure 7: Sentence-function distributions around action commitment for 1,004 decisions and 141,377 sentences. The top panel compares function frequencies before and after commitment. The bottom panel shows their local evolution around the commitment position and compares the observed change with a position-matched placebo. Meta-checking should be interpreted as exploratory because its annotation precision is low.
Measure
Before
At or after
Change [95% CI]
Belief entropy (bits)
0.392
0.336
−0.056[−0.105,−0.009]
Report accuracy
82.8%
83.0%
+0.2[0.0,0.6] pp
Appendix
Table 5: Changes in report-based beliefs around retrospective action commitment. Values are averaged within decisions before being aggregated across decisions. Intervals are 95% trajectory-clustered bootstrap intervals.
Transition
Events
Any report change
Δ error [95% CI]
Any action change
682
14 (2.1%)
0.0005[−0.0022,0.0027]
Optimal to suboptimal
161
6 (3.7%)
0.0000[−0.0065,0.0055]
Suboptimal to optimal
182
3 (1.6%)
−0.0018[−0.0080,0.0017]
Appendix
Table 6: Changes in report correctness at action transitions. “Any report change” counts events where the correctness of at least one of the six current-state reports changes. Error changes compare the preceding sentence with the transition sentence.
Cluster
True commitment
Matched control
Excess
C10
14.0%
5.9%
+8.2 pp
C9
11.6%
8.4%
+3.2 pp
C2
9.3%
6.8%
+2.5 pp
C5
15.1%
13.5%
+1.6 pp
C7
5.5%
4.5%
+1.0 pp
Appendix
Table 7: Frequency of selected activation clusters at true commitment sentences and position-matched control sentences from the same decisions.
Cluster
Action change
Optimal to suboptimal
Suboptimal to optimal
C6
1.65×
2.07×
1.77×
C10
1.64×
1.48×
1.59×
C9
1.41×
1.45×
1.44×
C5
1.33×
1.26×
1.35×
Appendix
Table 8: Enrichment of sentence-level activation clusters at changes in preferred action and preferred-action optimality relative to within-decision shuffled-event controls.
Figure 8: Whole-chain reasoning-text patching. The prompt from decision A is paired with a reasoning prefix from decision B , whose emitted action differs from that of A . Across early, middle, and late cuts, the forced-action distribution preferentially follows the transplanted reasoning rather than the original prompt. The experiment uses 396 mismatched pairs and 396 congruent controls.
Figure 9: Single-sentence intervention results across 240 reasoning prefixes and 720 independent forced-action evaluations. Left: probability that the forced top action becomes the injected runner-up action as a function of current action confidence. Right: mean probability movement of the injected action under counter-evidence, neutral, and supportive interventions.
Figure 10: Belief–action decomposition: worked example and empirical summary. (A–B) A constructed DoorKey example assigns belief probability to nine positions, all with no key and a locked door; blank belief cells have zero probability. The green map assigns values to all 39 non-wall positions. For this illustration only, g(x,y,h,d)=v(x,y) across valid states; the door-cell value applies when the door is open. With Q(s,a)=g(TG(s,a)) , the emitted RIGHT action has expected value 2.47 , the best single action UP has value 3.95 , and choosing after learning the state yields 5.34 . Thus the belief–action gap is R=1.48 , the value of perfect state information is I=1.39 , and expected local regret is Δ=R+I=2.87 . RIGHT is blocked at the outlined actual position. These beliefs, values, and action are illustrative, not fitted observations or task rewards. (C) Empirical endpoint means across all 3,194 decisions ( Table 1 ), using representation-based beliefs and agent-supervised value models fitted separately before and after reasoning. R decreases, while I and decoded-belief entropy H increase.
LLM decision-making studies how language models choose, act, seek information, and allocate resources under goals, uncertainty, constraints, and social pressures.
Task-oriented dialogue requires maintaining and updating information across turns, yet language models expose no explicit belief-state object. We study how conversational state is represented, updated, and used inside eight instruction-tuned language models from four families on MultiWOZ and SGD. Structure and values separate: which domains, slots, and requests are active is linearly readable just before the model acts, whereas exact values are far more readable where the user stated them. After a user changes a value, both values remain accessible at their mentions, and causal interventions show that both continue to influence the model's action. In natural closed-loop interaction, query failures separate into cases of weak structural support, incorrect value resolution, and failure to deploy otherwise-supported constraints, with targeted interventions producing systematically different repair behavior across these cases. These findings motivate a state-action controller that starts from the base model action and selectively edits it using structural readouts, without requiring a complete predicted belief state as an intermediate representation. On held-out MultiWOZ interaction across five models, it raises the base model exact-query accuracy from .318 to .621 and task success from .272 to .371 at negligible added cost. Overall, reliable interaction requires not only retaining conversational information, but resolving which available constraints currently apply and ensuring that they govern action.
Language-model agents often reason before they act, and the value of monitoring that reasoning depends on how it relates to their actions. We ask how an agent's actions depend on its reasoning when the correct action depends on a changing state. In a small file-system task, background events announce which file's lock has been toggled but not its new value. The agent therefore has to reconstruct the current locks from its history before it can move files safely. With Qwen3-8B and Qwen3.5-9B, we separate two roles of reasoning: generating it before each action, and retaining earlier reasoning in the prompt. Most of the reduction in lock-violating commands comes from reasoning written at the moment of action. Retaining earlier reasoning adds little to rule-following, although it helps one model complete the task. The benefit depends on what the reasoning says. Filler text, reasoning written for another history and an accurate table of the state do not reproduce it. When we give the agent reasoning written for a history that differs in one lock value, the agent acts on that reasoning even against its own history. Without reasoning, linear probes recover the state from the model's activations about as well as a text classifier recovers it from the transcript. However, decoding accuracy does not predict which interventions change the agent's actions. In this setting, the benefit of reasoning for rule-following comes mainly from generating it at each step. The state that can be decoded without reasoning is a limited guide to how the agent acts.
Linear probes and activation steering have uncovered that vision-language models (VLMs) internally represent mental states such as agents' beliefs, knowledge, and intentions. However, it is unclear whether and how these representations are used by downstream predictions along these axes. To close this gap, we introduce Cross-Axis Routing Diagnostic (CARD), which steers activations along one axis while measuring the response of a different axis's prediction. Applied to open-weight VLMs on Relay Chain -- a new cooperative grid-world benchmark we propose -- we diagnose a critical routing failure: models fail to incorporate belief representations into their next action prediction, effectively leaving valuable information about their partners unused.
Souptik Kumar Majumdar, Fabian Kögel, Andreas Bulling